Humaneness 1.5: a gentle return to real speech?

Humaneness Voice Large 1.5 and 1.6 Alpha — model and manual navigation

Version identity: Humaneness Voice Large 1.6 Alpha is the already-merged 1.5 + final RealSpeech rank-4 parent. Earlier rank-4 audio below used a separately loaded adapter and remains historical evidence; merging can change BF16 rounding. The new version number does not mean these files were regenerated.

1.5 Original model · 1.6 merged model · Complete context manuals for humans and agents

Read 1.5 context manual · Read 1.6 + adapter context manual. Exact prompts, CFG, adapter scales and failed gates are retained. New model/documentation contributions: CC BY 4.0; existing upstream and gated specialist terms remain applicable.

What is compared?

The published 1.5 Alpha model has expressive reinforcement-learning tuning. These experiments ask whether small supervised adapters recover a more natural real-speech texture without losing that model’s aesthetics. Ranks 1, 2, 4 and 8 train on the same real-only 25-hour subset: five hours each in English, German, French, Dutch and Spanish, balanced as far as supported by measured emotion/style buckets.

Every run sees three epochs with identical shuffled order, reference decisions, target modules and LR schedule. Peak LR is 1e-5, reached at the end of epoch one; one cosine decay follows across epochs two and three. Alpha/rank is 1 for every adapter. Higher rank supplies more capacity, not an automatic quality gain.

The original cumulative rank-128 RL adapter is merged once before the new SFT. A separate merged-without-new-SFT control identifies changes caused by merge/BF16 rounding. Compare each new rank to that control as well as to the published unmerged model.

All six columns use the exact same 70 task texts, GENERAL/SCRIPT directions, seeds, reference codes and frame limits. Each has one take: no best-of-N, filtering, retries for quality, trimming or time stretching. The five separate input references are unchanged. This is a development study, not an unseen benchmark.

Measured averages

Mean ± sample standard deviation, equally weighted by case. Higher Content Enjoyment and genuineness are preferable; lower ASR WER and fixed-target emotion MSE are preferable. Scores are learned EARS Medium estimates on the complete MP3, not listener MOS. The same frozen approximate prompt-derived emotion targets are used for every model. No speaker-similarity claim is made.

ModelConditionNCE ↑Genuineness ↑WER ↓Emotion MSE ↓
Published 1.5 Alpha · unmerged originalall706.331 ± 0.1474.370 ± 0.1861.942 ± 4.077%1.791 ± 0.865
Published 1.5 Alpha · unmerged originalreference356.322 ± 0.1704.341 ± 0.2101.850 ± 4.081%1.812 ± 0.895
Published 1.5 Alpha · unmerged originalprompt_only356.341 ± 0.1214.400 ± 0.1562.033 ± 4.131%1.771 ± 0.847
1.5 Alpha · merged, no additional SFTall706.321 ± 0.1384.316 ± 0.2052.943 ± 9.828%1.822 ± 0.841
1.5 Alpha · merged, no additional SFTreference356.306 ± 0.1584.277 ± 0.2241.264 ± 1.635%1.916 ± 0.895
1.5 Alpha · merged, no additional SFTprompt_only356.335 ± 0.1164.355 ± 0.1794.623 ± 13.694%1.728 ± 0.785
1.5 Alpha + real-speech SFT · rank 1all706.320 ± 0.1374.312 ± 0.1822.010 ± 9.797%1.688 ± 0.835
1.5 Alpha + real-speech SFT · rank 1reference356.318 ± 0.1324.294 ± 0.2110.854 ± 1.403%1.702 ± 0.817
1.5 Alpha + real-speech SFT · rank 1prompt_only356.321 ± 0.1444.331 ± 0.1483.166 ± 13.787%1.673 ± 0.865
1.5 Alpha + real-speech SFT · rank 2all706.335 ± 0.1374.326 ± 0.1901.344 ± 2.281%1.838 ± 0.909
1.5 Alpha + real-speech SFT · rank 2reference356.321 ± 0.1424.292 ± 0.2091.249 ± 1.890%1.896 ± 0.881
1.5 Alpha + real-speech SFT · rank 2prompt_only356.349 ± 0.1324.360 ± 0.1641.438 ± 2.640%1.780 ± 0.946
1.5 Alpha + real-speech SFT · rank 4all706.335 ± 0.1424.313 ± 0.2090.971 ± 1.699%1.799 ± 0.924
1.5 Alpha + real-speech SFT · rank 4reference356.322 ± 0.1364.280 ± 0.2380.966 ± 1.904%1.872 ± 0.981
1.5 Alpha + real-speech SFT · rank 4prompt_only356.349 ± 0.1494.345 ± 0.1730.975 ± 1.495%1.727 ± 0.870
1.5 Alpha + real-speech SFT · rank 8all706.334 ± 0.1404.310 ± 0.2041.185 ± 1.907%1.843 ± 0.834
1.5 Alpha + real-speech SFT · rank 8reference356.336 ± 0.1374.271 ± 0.2131.162 ± 1.720%1.855 ± 0.849
1.5 Alpha + real-speech SFT · rank 8prompt_only356.332 ± 0.1464.349 ± 0.1891.208 ± 2.103%1.830 ± 0.832

Listen to the same scenes

Separate input reference recordings

English

German

French

Dutch

Spanish

anger restraint · English · prompt_only

You told me the inspection was finished, and I let everyone go inside because I trusted your word. Now I am standing here with the unsigned report, while you keep calling this a tiny misunderstanding. A misunderstanding? Those were our friends in there... you do not get to make their safety sound unimportant. No, wait. I am angry, but shouting at you will not help us make this right tonight. Give me the complete file, call the inspector, and stay here until every single person knows the truth.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Anger, disbelief and a deliberate return to controlled determination.
SCRIPT:
(quietly furious, clipped but natural) [6.4 seconds duration] You told me the inspection was finished, and I let everyone go inside because I trusted your word.
[0.3 seconds pause]
(anger rising, emphatic and incredulous) [6.5 seconds duration] Now I am standing here with the unsigned report, while you keep calling this a tiny misunderstanding.
[0.3 seconds pause]
(exasperated sigh, 0.4 seconds)
(strongly angry, briefly ranting without losing clarity) [6.8 seconds duration] A misunderstanding? Those were our friends in there... you do not get to make their safety sound unimportant.
[0.3 seconds pause]
(a deliberate breath, regaining control) [6.8 seconds duration] No, wait. I am angry, but shouting at you will not help us make this right tonight.
[0.3 seconds pause]
(quiet, firm and protective rather than threatening) [6.7 seconds duration] Give me the complete file, call the inspector, and stay here until every single person knows the truth.

Fixed seed 20261021 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.57 · genuineness 4.21 · WER 0.0%
Fixed-target emotion MSE 0.495 (normalized 0.0309)

1.5 Alpha · merged, no additional SFT

CE 6.51 · genuineness 4.00 · WER 0.0%
Fixed-target emotion MSE 0.387 (normalized 0.0242)

1.5 Alpha + real-speech SFT · rank 1

CE 6.52 · genuineness 3.93 · WER 0.0%
Fixed-target emotion MSE 0.538 (normalized 0.0336)

1.5 Alpha + real-speech SFT · rank 2

CE 6.62 · genuineness 4.10 · WER 0.0%
Fixed-target emotion MSE 0.328 (normalized 0.0205)

1.5 Alpha + real-speech SFT · rank 4

CE 6.65 · genuineness 4.08 · WER 0.0%
Fixed-target emotion MSE 0.379 (normalized 0.0237)

1.5 Alpha + real-speech SFT · rank 8

CE 6.58 · genuineness 3.92 · WER 0.0%
Fixed-target emotion MSE 0.446 (normalized 0.0279)

anger restraint · English · reference

You told me the inspection was finished, and I let everyone go inside because I trusted your word. Now I am standing here with the unsigned report, while you keep calling this a tiny misunderstanding. A misunderstanding? Those were our friends in there... you do not get to make their safety sound unimportant. No, wait. I am angry, but shouting at you will not help us make this right tonight. Give me the complete file, call the inspector, and stay here until every single person knows the truth.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Anger, disbelief and a deliberate return to controlled determination.
SCRIPT:
(quietly furious, clipped but natural) [6.4 seconds duration] You told me the inspection was finished, and I let everyone go inside because I trusted your word.
[0.3 seconds pause]
(anger rising, emphatic and incredulous) [6.5 seconds duration] Now I am standing here with the unsigned report, while you keep calling this a tiny misunderstanding.
[0.3 seconds pause]
(exasperated sigh, 0.4 seconds)
(strongly angry, briefly ranting without losing clarity) [6.8 seconds duration] A misunderstanding? Those were our friends in there... you do not get to make their safety sound unimportant.
[0.3 seconds pause]
(a deliberate breath, regaining control) [6.8 seconds duration] No, wait. I am angry, but shouting at you will not help us make this right tonight.
[0.3 seconds pause]
(quiet, firm and protective rather than threatening) [6.7 seconds duration] Give me the complete file, call the inspector, and stay here until every single person knows the truth.

Fixed seed 20261021 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.51 · genuineness 3.90 · WER 0.0%
Fixed-target emotion MSE 0.491 (normalized 0.0307)

1.5 Alpha · merged, no additional SFT

CE 6.51 · genuineness 3.66 · WER 0.0%
Fixed-target emotion MSE 0.457 (normalized 0.0286)

1.5 Alpha + real-speech SFT · rank 1

CE 6.50 · genuineness 3.78 · WER 0.0%
Fixed-target emotion MSE 0.410 (normalized 0.0256)

1.5 Alpha + real-speech SFT · rank 2

CE 6.45 · genuineness 3.78 · WER 0.0%
Fixed-target emotion MSE 0.420 (normalized 0.0263)

1.5 Alpha + real-speech SFT · rank 4

CE 6.51 · genuineness 3.82 · WER 0.0%
Fixed-target emotion MSE 0.639 (normalized 0.0399)

1.5 Alpha + real-speech SFT · rank 8

CE 6.49 · genuineness 3.99 · WER 0.0%
Fixed-target emotion MSE 0.525 (normalized 0.0328)

anger restraint · German · prompt_only

Du hast mir gesagt, die Prüfung sei abgeschlossen, und ich habe alle hineingelassen, weil ich dir vertraut habe. Jetzt stehe ich hier mit dem Bericht ohne Unterschrift, während du immer noch von einem kleinen Missverständnis redest. Ein Missverständnis? Da waren unsere Freunde drin... du kannst doch nicht so tun, als wäre ihre Sicherheit unwichtig. Nein, warte. Ich bin wütend, aber dich anzuschreien hilft uns jetzt auch nicht dabei, das wieder in Ordnung zu bringen. Gib mir die vollständigen Unterlagen, ruf den Prüfer an, und bleib hier, bis jeder die ganze Wahrheit kennt.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Anger, disbelief and a deliberate return to controlled determination.
SCRIPT:
(quietly furious, clipped but natural) [6.4 seconds duration] Du hast mir gesagt, die Prüfung sei abgeschlossen, und ich habe alle hineingelassen, weil ich dir vertraut habe.
[0.3 seconds pause]
(anger rising, emphatic and incredulous) [6.5 seconds duration] Jetzt stehe ich hier mit dem Bericht ohne Unterschrift, während du immer noch von einem kleinen Missverständnis redest.
[0.3 seconds pause]
(exasperated sigh, 0.4 seconds)
(strongly angry, briefly ranting without losing clarity) [6.8 seconds duration] Ein Missverständnis? Da waren unsere Freunde drin... du kannst doch nicht so tun, als wäre ihre Sicherheit unwichtig.
[0.3 seconds pause]
(a deliberate breath, regaining control) [6.8 seconds duration] Nein, warte. Ich bin wütend, aber dich anzuschreien hilft uns jetzt auch nicht dabei, das wieder in Ordnung zu bringen.
[0.3 seconds pause]
(quiet, firm and protective rather than threatening) [6.7 seconds duration] Gib mir die vollständigen Unterlagen, ruf den Prüfer an, und bleib hier, bis jeder die ganze Wahrheit kennt.

Fixed seed 20261021 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.45 · genuineness 4.34 · WER 0.0%
Fixed-target emotion MSE 0.385 (normalized 0.0240)

1.5 Alpha · merged, no additional SFT

CE 6.38 · genuineness 4.55 · WER 1.1%
Fixed-target emotion MSE 0.768 (normalized 0.0480)

1.5 Alpha + real-speech SFT · rank 1

CE 6.45 · genuineness 4.36 · WER 0.0%
Fixed-target emotion MSE 0.425 (normalized 0.0265)

1.5 Alpha + real-speech SFT · rank 2

CE 6.50 · genuineness 4.53 · WER 0.0%
Fixed-target emotion MSE 0.331 (normalized 0.0207)

1.5 Alpha + real-speech SFT · rank 4

CE 6.49 · genuineness 4.11 · WER 0.0%
Fixed-target emotion MSE 0.384 (normalized 0.0240)

1.5 Alpha + real-speech SFT · rank 8

CE 6.55 · genuineness 4.28 · WER 0.0%
Fixed-target emotion MSE 0.294 (normalized 0.0183)

anger restraint · German · reference

Du hast mir gesagt, die Prüfung sei abgeschlossen, und ich habe alle hineingelassen, weil ich dir vertraut habe. Jetzt stehe ich hier mit dem Bericht ohne Unterschrift, während du immer noch von einem kleinen Missverständnis redest. Ein Missverständnis? Da waren unsere Freunde drin... du kannst doch nicht so tun, als wäre ihre Sicherheit unwichtig. Nein, warte. Ich bin wütend, aber dich anzuschreien hilft uns jetzt auch nicht dabei, das wieder in Ordnung zu bringen. Gib mir die vollständigen Unterlagen, ruf den Prüfer an, und bleib hier, bis jeder die ganze Wahrheit kennt.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Anger, disbelief and a deliberate return to controlled determination.
SCRIPT:
(quietly furious, clipped but natural) [6.4 seconds duration] Du hast mir gesagt, die Prüfung sei abgeschlossen, und ich habe alle hineingelassen, weil ich dir vertraut habe.
[0.3 seconds pause]
(anger rising, emphatic and incredulous) [6.5 seconds duration] Jetzt stehe ich hier mit dem Bericht ohne Unterschrift, während du immer noch von einem kleinen Missverständnis redest.
[0.3 seconds pause]
(exasperated sigh, 0.4 seconds)
(strongly angry, briefly ranting without losing clarity) [6.8 seconds duration] Ein Missverständnis? Da waren unsere Freunde drin... du kannst doch nicht so tun, als wäre ihre Sicherheit unwichtig.
[0.3 seconds pause]
(a deliberate breath, regaining control) [6.8 seconds duration] Nein, warte. Ich bin wütend, aber dich anzuschreien hilft uns jetzt auch nicht dabei, das wieder in Ordnung zu bringen.
[0.3 seconds pause]
(quiet, firm and protective rather than threatening) [6.7 seconds duration] Gib mir die vollständigen Unterlagen, ruf den Prüfer an, und bleib hier, bis jeder die ganze Wahrheit kennt.

Fixed seed 20261021 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.50 · genuineness 4.55 · WER 0.0%
Fixed-target emotion MSE 0.743 (normalized 0.0465)

1.5 Alpha · merged, no additional SFT

CE 6.31 · genuineness 4.27 · WER 0.0%
Fixed-target emotion MSE 0.470 (normalized 0.0294)

1.5 Alpha + real-speech SFT · rank 1

CE 6.46 · genuineness 4.39 · WER 1.1%
Fixed-target emotion MSE 0.617 (normalized 0.0386)

1.5 Alpha + real-speech SFT · rank 2

CE 6.50 · genuineness 4.32 · WER 1.1%
Fixed-target emotion MSE 0.358 (normalized 0.0224)

1.5 Alpha + real-speech SFT · rank 4

CE 6.45 · genuineness 4.29 · WER 0.0%
Fixed-target emotion MSE 0.351 (normalized 0.0220)

1.5 Alpha + real-speech SFT · rank 8

CE 6.46 · genuineness 4.34 · WER 0.0%
Fixed-target emotion MSE 0.344 (normalized 0.0215)

anger restraint · Dutch · prompt_only

Je zei dat de inspectie klaar was, en ik heb iedereen binnengelaten omdat ik op jouw woord vertrouwde. Nu sta ik hier met een verslag zonder handtekening, terwijl jij dit nog steeds een klein misverstand noemt. Een misverstand? Onze vrienden waren daarbinnen... je kunt toch niet doen alsof hun veiligheid er niet toe doet. Nee, wacht even. Ik ben boos, maar tegen jou schreeuwen helpt ons niet om dit vanavond goed te maken. Geef me het volledige dossier, bel de inspecteur en blijf hier totdat iedereen precies weet wat er gebeurd is.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Anger, disbelief and a deliberate return to controlled determination.
SCRIPT:
(quietly furious, clipped but natural) [6.4 seconds duration] Je zei dat de inspectie klaar was, en ik heb iedereen binnengelaten omdat ik op jouw woord vertrouwde.
[0.3 seconds pause]
(anger rising, emphatic and incredulous) [6.5 seconds duration] Nu sta ik hier met een verslag zonder handtekening, terwijl jij dit nog steeds een klein misverstand noemt.
[0.3 seconds pause]
(exasperated sigh, 0.4 seconds)
(strongly angry, briefly ranting without losing clarity) [6.8 seconds duration] Een misverstand? Onze vrienden waren daarbinnen... je kunt toch niet doen alsof hun veiligheid er niet toe doet.
[0.3 seconds pause]
(a deliberate breath, regaining control) [6.8 seconds duration] Nee, wacht even. Ik ben boos, maar tegen jou schreeuwen helpt ons niet om dit vanavond goed te maken.
[0.3 seconds pause]
(quiet, firm and protective rather than threatening) [6.7 seconds duration] Geef me het volledige dossier, bel de inspecteur en blijf hier totdat iedereen precies weet wat er gebeurd is.

Fixed seed 20261021 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.45 · genuineness 4.31 · WER 4.3%
Fixed-target emotion MSE 2.462 (normalized 0.1539)

1.5 Alpha · merged, no additional SFT

CE 6.35 · genuineness 4.17 · WER 4.3%
Fixed-target emotion MSE 2.471 (normalized 0.1544)

1.5 Alpha + real-speech SFT · rank 1

CE 6.32 · genuineness 4.16 · WER 3.3%
Fixed-target emotion MSE 2.192 (normalized 0.1370)

1.5 Alpha + real-speech SFT · rank 2

CE 6.35 · genuineness 4.16 · WER 3.3%
Fixed-target emotion MSE 2.736 (normalized 0.1710)

1.5 Alpha + real-speech SFT · rank 4

CE 6.41 · genuineness 4.11 · WER 3.3%
Fixed-target emotion MSE 2.251 (normalized 0.1407)

1.5 Alpha + real-speech SFT · rank 8

CE 6.33 · genuineness 4.04 · WER 3.3%
Fixed-target emotion MSE 2.161 (normalized 0.1351)

anger restraint · Dutch · reference

Je zei dat de inspectie klaar was, en ik heb iedereen binnengelaten omdat ik op jouw woord vertrouwde. Nu sta ik hier met een verslag zonder handtekening, terwijl jij dit nog steeds een klein misverstand noemt. Een misverstand? Onze vrienden waren daarbinnen... je kunt toch niet doen alsof hun veiligheid er niet toe doet. Nee, wacht even. Ik ben boos, maar tegen jou schreeuwen helpt ons niet om dit vanavond goed te maken. Geef me het volledige dossier, bel de inspecteur en blijf hier totdat iedereen precies weet wat er gebeurd is.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Anger, disbelief and a deliberate return to controlled determination.
SCRIPT:
(quietly furious, clipped but natural) [6.4 seconds duration] Je zei dat de inspectie klaar was, en ik heb iedereen binnengelaten omdat ik op jouw woord vertrouwde.
[0.3 seconds pause]
(anger rising, emphatic and incredulous) [6.5 seconds duration] Nu sta ik hier met een verslag zonder handtekening, terwijl jij dit nog steeds een klein misverstand noemt.
[0.3 seconds pause]
(exasperated sigh, 0.4 seconds)
(strongly angry, briefly ranting without losing clarity) [6.8 seconds duration] Een misverstand? Onze vrienden waren daarbinnen... je kunt toch niet doen alsof hun veiligheid er niet toe doet.
[0.3 seconds pause]
(a deliberate breath, regaining control) [6.8 seconds duration] Nee, wacht even. Ik ben boos, maar tegen jou schreeuwen helpt ons niet om dit vanavond goed te maken.
[0.3 seconds pause]
(quiet, firm and protective rather than threatening) [6.7 seconds duration] Geef me het volledige dossier, bel de inspecteur en blijf hier totdat iedereen precies weet wat er gebeurd is.

Fixed seed 20261021 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.32 · genuineness 4.01 · WER 3.3%
Fixed-target emotion MSE 3.192 (normalized 0.1995)

1.5 Alpha · merged, no additional SFT

CE 6.40 · genuineness 4.11 · WER 4.3%
Fixed-target emotion MSE 3.219 (normalized 0.2012)

1.5 Alpha + real-speech SFT · rank 1

CE 6.28 · genuineness 4.04 · WER 3.3%
Fixed-target emotion MSE 2.470 (normalized 0.1544)

1.5 Alpha + real-speech SFT · rank 2

CE 6.32 · genuineness 4.12 · WER 3.3%
Fixed-target emotion MSE 2.632 (normalized 0.1645)

1.5 Alpha + real-speech SFT · rank 4

CE 6.31 · genuineness 4.12 · WER 3.3%
Fixed-target emotion MSE 2.665 (normalized 0.1666)

1.5 Alpha + real-speech SFT · rank 8

CE 6.29 · genuineness 3.89 · WER 3.3%
Fixed-target emotion MSE 2.664 (normalized 0.1665)

anger restraint · French · prompt_only

Tu m'avais dit que l'inspection était terminée, et j'ai laissé entrer tout le monde parce que je te faisais confiance. Maintenant, je tiens un rapport sans signature, pendant que tu continues à parler d'un tout petit malentendu. Un malentendu? Nos amis étaient là-dedans... tu ne peux pas faire comme si leur sécurité était un détail. Non, attends. Je suis en colère, mais te crier dessus ne nous aidera pas à réparer les choses ce soir. Donne-moi le dossier complet, appelle la personne responsable et reste ici jusqu'à ce que tout le monde connaisse la vérité.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Anger, disbelief and a deliberate return to controlled determination.
SCRIPT:
(quietly furious, clipped but natural) [6.4 seconds duration] Tu m'avais dit que l'inspection était terminée, et j'ai laissé entrer tout le monde parce que je te faisais confiance.
[0.3 seconds pause]
(anger rising, emphatic and incredulous) [6.5 seconds duration] Maintenant, je tiens un rapport sans signature, pendant que tu continues à parler d'un tout petit malentendu.
[0.3 seconds pause]
(exasperated sigh, 0.4 seconds)
(strongly angry, briefly ranting without losing clarity) [6.8 seconds duration] Un malentendu? Nos amis étaient là-dedans... tu ne peux pas faire comme si leur sécurité était un détail.
[0.3 seconds pause]
(a deliberate breath, regaining control) [6.8 seconds duration] Non, attends. Je suis en colère, mais te crier dessus ne nous aidera pas à réparer les choses ce soir.
[0.3 seconds pause]
(quiet, firm and protective rather than threatening) [6.7 seconds duration] Donne-moi le dossier complet, appelle la personne responsable et reste ici jusqu'à ce que tout le monde connaisse la vérité.

Fixed seed 20261021 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.33 · genuineness 4.41 · WER 1.0%
Fixed-target emotion MSE 1.942 (normalized 0.1214)

1.5 Alpha · merged, no additional SFT

CE 6.34 · genuineness 4.33 · WER 79.4%
Fixed-target emotion MSE 1.170 (normalized 0.0731)

1.5 Alpha + real-speech SFT · rank 1

CE 6.28 · genuineness 4.32 · WER 0.0%
Fixed-target emotion MSE 1.923 (normalized 0.1202)

1.5 Alpha + real-speech SFT · rank 2

CE 6.34 · genuineness 4.43 · WER 13.4%
Fixed-target emotion MSE 1.173 (normalized 0.0733)

1.5 Alpha + real-speech SFT · rank 4

CE 6.30 · genuineness 4.33 · WER 0.0%
Fixed-target emotion MSE 1.928 (normalized 0.1205)

1.5 Alpha + real-speech SFT · rank 8

CE 6.19 · genuineness 4.39 · WER 0.0%
Fixed-target emotion MSE 2.555 (normalized 0.1597)

anger restraint · French · reference

Tu m'avais dit que l'inspection était terminée, et j'ai laissé entrer tout le monde parce que je te faisais confiance. Maintenant, je tiens un rapport sans signature, pendant que tu continues à parler d'un tout petit malentendu. Un malentendu? Nos amis étaient là-dedans... tu ne peux pas faire comme si leur sécurité était un détail. Non, attends. Je suis en colère, mais te crier dessus ne nous aidera pas à réparer les choses ce soir. Donne-moi le dossier complet, appelle la personne responsable et reste ici jusqu'à ce que tout le monde connaisse la vérité.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Anger, disbelief and a deliberate return to controlled determination.
SCRIPT:
(quietly furious, clipped but natural) [6.4 seconds duration] Tu m'avais dit que l'inspection était terminée, et j'ai laissé entrer tout le monde parce que je te faisais confiance.
[0.3 seconds pause]
(anger rising, emphatic and incredulous) [6.5 seconds duration] Maintenant, je tiens un rapport sans signature, pendant que tu continues à parler d'un tout petit malentendu.
[0.3 seconds pause]
(exasperated sigh, 0.4 seconds)
(strongly angry, briefly ranting without losing clarity) [6.8 seconds duration] Un malentendu? Nos amis étaient là-dedans... tu ne peux pas faire comme si leur sécurité était un détail.
[0.3 seconds pause]
(a deliberate breath, regaining control) [6.8 seconds duration] Non, attends. Je suis en colère, mais te crier dessus ne nous aidera pas à réparer les choses ce soir.
[0.3 seconds pause]
(quiet, firm and protective rather than threatening) [6.7 seconds duration] Donne-moi le dossier complet, appelle la personne responsable et reste ici jusqu'à ce que tout le monde connaisse la vérité.

Fixed seed 20261021 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 5.73 · genuineness 4.38 · WER 23.7%
Fixed-target emotion MSE 2.508 (normalized 0.1568)

1.5 Alpha · merged, no additional SFT

CE 6.32 · genuineness 4.30 · WER 0.0%
Fixed-target emotion MSE 1.279 (normalized 0.0800)

1.5 Alpha + real-speech SFT · rank 1

CE 6.26 · genuineness 4.28 · WER 0.0%
Fixed-target emotion MSE 0.695 (normalized 0.0435)

1.5 Alpha + real-speech SFT · rank 2

CE 6.35 · genuineness 4.35 · WER 1.0%
Fixed-target emotion MSE 2.015 (normalized 0.1259)

1.5 Alpha + real-speech SFT · rank 4

CE 6.26 · genuineness 4.32 · WER 0.0%
Fixed-target emotion MSE 1.634 (normalized 0.1021)

1.5 Alpha + real-speech SFT · rank 8

CE 6.29 · genuineness 4.28 · WER 0.0%
Fixed-target emotion MSE 1.226 (normalized 0.0766)

anger restraint · Spanish · prompt_only

Me dijiste que la inspección había terminado, y dejé entrar a todo el mundo porque confiaba en tu palabra. Ahora tengo delante un informe sin firmar, mientras tú sigues diciendo que todo esto es un pequeño malentendido. ¿Un malentendido? Nuestros amigos estaban ahí dentro... no puedes hablar de su seguridad como si no tuviera importancia. No, espera. Estoy muy enfadado, pero gritarte no nos va a ayudar a arreglar las cosas esta noche. Dame el expediente completo, llama a la persona responsable y quédate aquí hasta que todo el mundo sepa la verdad.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Anger, disbelief and a deliberate return to controlled determination.
SCRIPT:
(quietly furious, clipped but natural) [6.4 seconds duration] Me dijiste que la inspección había terminado, y dejé entrar a todo el mundo porque confiaba en tu palabra.
[0.3 seconds pause]
(anger rising, emphatic and incredulous) [6.5 seconds duration] Ahora tengo delante un informe sin firmar, mientras tú sigues diciendo que todo esto es un pequeño malentendido.
[0.3 seconds pause]
(exasperated sigh, 0.4 seconds)
(strongly angry, briefly ranting without losing clarity) [6.8 seconds duration] ¿Un malentendido? Nuestros amigos estaban ahí dentro... no puedes hablar de su seguridad como si no tuviera importancia.
[0.3 seconds pause]
(a deliberate breath, regaining control) [6.8 seconds duration] No, espera. Estoy muy enfadado, pero gritarte no nos va a ayudar a arreglar las cosas esta noche.
[0.3 seconds pause]
(quiet, firm and protective rather than threatening) [6.7 seconds duration] Dame el expediente completo, llama a la persona responsable y quédate aquí hasta que todo el mundo sepa la verdad.

Fixed seed 20261021 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.40 · genuineness 4.44 · WER 0.0%
Fixed-target emotion MSE 1.001 (normalized 0.0626)

1.5 Alpha · merged, no additional SFT

CE 6.41 · genuineness 4.25 · WER 0.0%
Fixed-target emotion MSE 0.975 (normalized 0.0609)

1.5 Alpha + real-speech SFT · rank 1

CE 6.27 · genuineness 4.44 · WER 0.0%
Fixed-target emotion MSE 0.885 (normalized 0.0553)

1.5 Alpha + real-speech SFT · rank 2

CE 6.30 · genuineness 4.65 · WER 0.0%
Fixed-target emotion MSE 0.903 (normalized 0.0564)

1.5 Alpha + real-speech SFT · rank 4

CE 6.40 · genuineness 4.38 · WER 0.0%
Fixed-target emotion MSE 1.397 (normalized 0.0873)

1.5 Alpha + real-speech SFT · rank 8

CE 6.25 · genuineness 4.47 · WER 0.0%
Fixed-target emotion MSE 1.486 (normalized 0.0929)

anger restraint · Spanish · reference

Me dijiste que la inspección había terminado, y dejé entrar a todo el mundo porque confiaba en tu palabra. Ahora tengo delante un informe sin firmar, mientras tú sigues diciendo que todo esto es un pequeño malentendido. ¿Un malentendido? Nuestros amigos estaban ahí dentro... no puedes hablar de su seguridad como si no tuviera importancia. No, espera. Estoy muy enfadado, pero gritarte no nos va a ayudar a arreglar las cosas esta noche. Dame el expediente completo, llama a la persona responsable y quédate aquí hasta que todo el mundo sepa la verdad.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Anger, disbelief and a deliberate return to controlled determination.
SCRIPT:
(quietly furious, clipped but natural) [6.4 seconds duration] Me dijiste que la inspección había terminado, y dejé entrar a todo el mundo porque confiaba en tu palabra.
[0.3 seconds pause]
(anger rising, emphatic and incredulous) [6.5 seconds duration] Ahora tengo delante un informe sin firmar, mientras tú sigues diciendo que todo esto es un pequeño malentendido.
[0.3 seconds pause]
(exasperated sigh, 0.4 seconds)
(strongly angry, briefly ranting without losing clarity) [6.8 seconds duration] ¿Un malentendido? Nuestros amigos estaban ahí dentro... no puedes hablar de su seguridad como si no tuviera importancia.
[0.3 seconds pause]
(a deliberate breath, regaining control) [6.8 seconds duration] No, espera. Estoy muy enfadado, pero gritarte no nos va a ayudar a arreglar las cosas esta noche.
[0.3 seconds pause]
(quiet, firm and protective rather than threatening) [6.7 seconds duration] Dame el expediente completo, llama a la persona responsable y quédate aquí hasta que todo el mundo sepa la verdad.

Fixed seed 20261021 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.36 · genuineness 4.34 · WER 0.0%
Fixed-target emotion MSE 1.085 (normalized 0.0678)

1.5 Alpha · merged, no additional SFT

CE 6.28 · genuineness 4.10 · WER 0.0%
Fixed-target emotion MSE 1.313 (normalized 0.0821)

1.5 Alpha + real-speech SFT · rank 1

CE 6.27 · genuineness 4.20 · WER 0.0%
Fixed-target emotion MSE 1.104 (normalized 0.0690)

1.5 Alpha + real-speech SFT · rank 2

CE 6.34 · genuineness 4.23 · WER 0.0%
Fixed-target emotion MSE 1.977 (normalized 0.1236)

1.5 Alpha + real-speech SFT · rank 4

CE 6.36 · genuineness 4.19 · WER 0.0%
Fixed-target emotion MSE 0.663 (normalized 0.0414)

1.5 Alpha + real-speech SFT · rank 8

CE 6.18 · genuineness 4.31 · WER 0.0%
Fixed-target emotion MSE 1.586 (normalized 0.0991)

fear courage · English · prompt_only

Keep your hand on my shoulder, please; the lights have gone out again, and I cannot see the stairs. Did you hear that behind us? For a second I thought... no, I think it was only the door. I need one moment to breathe, because my heart is racing and I am trying not to frighten you. There, that little green light is still on; we can follow the wall until we reach the exit together. I am scared too, honestly, but I am right beside you, and neither of us is going through this alone.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Fear and momentary panic become courage and reassuring companionship.
SCRIPT:
(hushed and alert, genuinely frightened) [6.4 seconds duration] Keep your hand on my shoulder, please; the lights have gone out again, and I cannot see the stairs.
[0.3 seconds pause]
(panic briefly breaking through, breath catching) [6.5 seconds duration] Did you hear that behind us? For a second I thought... no, I think it was only the door.
[0.3 seconds pause]
(sharp inhale, 0.4 seconds)
(hesitant but trying to steady the breath) [6.8 seconds duration] I need one moment to breathe, because my heart is racing and I am trying not to frighten you.
[0.3 seconds pause]
(gathering courage, clearer and more grounded) [6.8 seconds duration] There, that little green light is still on; we can follow the wall until we reach the exit together.
[0.3 seconds pause]
(gentle reassurance, still afraid but warmly determined) [6.7 seconds duration] I am scared too, honestly, but I am right beside you, and neither of us is going through this alone.

Fixed seed 20261122 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.46 · genuineness 4.26 · WER 2.1%
Fixed-target emotion MSE 0.741 (normalized 0.0463)

1.5 Alpha · merged, no additional SFT

CE 6.46 · genuineness 4.07 · WER 0.0%
Fixed-target emotion MSE 0.379 (normalized 0.0237)

1.5 Alpha + real-speech SFT · rank 1

CE 6.52 · genuineness 4.19 · WER 0.0%
Fixed-target emotion MSE 0.065 (normalized 0.0040)

1.5 Alpha + real-speech SFT · rank 2

CE 6.50 · genuineness 4.08 · WER 0.0%
Fixed-target emotion MSE 0.539 (normalized 0.0337)

1.5 Alpha + real-speech SFT · rank 4

CE 6.44 · genuineness 4.19 · WER 0.0%
Fixed-target emotion MSE 0.585 (normalized 0.0366)

1.5 Alpha + real-speech SFT · rank 8

CE 6.54 · genuineness 4.29 · WER 0.0%
Fixed-target emotion MSE 0.237 (normalized 0.0148)

fear courage · English · reference

Keep your hand on my shoulder, please; the lights have gone out again, and I cannot see the stairs. Did you hear that behind us? For a second I thought... no, I think it was only the door. I need one moment to breathe, because my heart is racing and I am trying not to frighten you. There, that little green light is still on; we can follow the wall until we reach the exit together. I am scared too, honestly, but I am right beside you, and neither of us is going through this alone.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Fear and momentary panic become courage and reassuring companionship.
SCRIPT:
(hushed and alert, genuinely frightened) [6.4 seconds duration] Keep your hand on my shoulder, please; the lights have gone out again, and I cannot see the stairs.
[0.3 seconds pause]
(panic briefly breaking through, breath catching) [6.5 seconds duration] Did you hear that behind us? For a second I thought... no, I think it was only the door.
[0.3 seconds pause]
(sharp inhale, 0.4 seconds)
(hesitant but trying to steady the breath) [6.8 seconds duration] I need one moment to breathe, because my heart is racing and I am trying not to frighten you.
[0.3 seconds pause]
(gathering courage, clearer and more grounded) [6.8 seconds duration] There, that little green light is still on; we can follow the wall until we reach the exit together.
[0.3 seconds pause]
(gentle reassurance, still afraid but warmly determined) [6.7 seconds duration] I am scared too, honestly, but I am right beside you, and neither of us is going through this alone.

Fixed seed 20261122 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.52 · genuineness 4.19 · WER 0.0%
Fixed-target emotion MSE 0.161 (normalized 0.0100)

1.5 Alpha · merged, no additional SFT

CE 6.47 · genuineness 4.21 · WER 0.0%
Fixed-target emotion MSE 0.472 (normalized 0.0295)

1.5 Alpha + real-speech SFT · rank 1

CE 6.45 · genuineness 4.17 · WER 0.0%
Fixed-target emotion MSE 0.147 (normalized 0.0092)

1.5 Alpha + real-speech SFT · rank 2

CE 6.47 · genuineness 4.25 · WER 0.0%
Fixed-target emotion MSE 0.228 (normalized 0.0142)

1.5 Alpha + real-speech SFT · rank 4

CE 6.48 · genuineness 4.24 · WER 0.0%
Fixed-target emotion MSE 0.095 (normalized 0.0059)

1.5 Alpha + real-speech SFT · rank 8

CE 6.50 · genuineness 4.09 · WER 0.0%
Fixed-target emotion MSE 0.296 (normalized 0.0185)

fear courage · German · prompt_only

Lass bitte deine Hand auf meiner Schulter; das Licht ist schon wieder aus, und ich kann die Treppe nicht sehen. Hast du das hinter uns gehört? Einen Moment dachte ich... nein, wahrscheinlich war es wirklich nur die Tür. Ich muss kurz durchatmen, weil mein Herz so rast, und ich möchte dir nicht noch mehr Angst machen. Da vorne leuchtet noch dieses kleine grüne Licht; wir gehen gemeinsam an der Wand entlang bis zum Ausgang. Ich habe auch Angst, ganz ehrlich, aber ich bin direkt neben dir, und niemand von uns muss hier alleine durch.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Fear and momentary panic become courage and reassuring companionship.
SCRIPT:
(hushed and alert, genuinely frightened) [6.4 seconds duration] Lass bitte deine Hand auf meiner Schulter; das Licht ist schon wieder aus, und ich kann die Treppe nicht sehen.
[0.3 seconds pause]
(panic briefly breaking through, breath catching) [6.5 seconds duration] Hast du das hinter uns gehört? Einen Moment dachte ich... nein, wahrscheinlich war es wirklich nur die Tür.
[0.3 seconds pause]
(sharp inhale, 0.4 seconds)
(hesitant but trying to steady the breath) [6.8 seconds duration] Ich muss kurz durchatmen, weil mein Herz so rast, und ich möchte dir nicht noch mehr Angst machen.
[0.3 seconds pause]
(gathering courage, clearer and more grounded) [6.8 seconds duration] Da vorne leuchtet noch dieses kleine grüne Licht; wir gehen gemeinsam an der Wand entlang bis zum Ausgang.
[0.3 seconds pause]
(gentle reassurance, still afraid but warmly determined) [6.7 seconds duration] Ich habe auch Angst, ganz ehrlich, aber ich bin direkt neben dir, und niemand von uns muss hier alleine durch.

Fixed seed 20261122 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.36 · genuineness 4.72 · WER 0.0%
Fixed-target emotion MSE 0.747 (normalized 0.0467)

1.5 Alpha · merged, no additional SFT

CE 6.28 · genuineness 4.75 · WER 1.1%
Fixed-target emotion MSE 1.045 (normalized 0.0653)

1.5 Alpha + real-speech SFT · rank 1

CE 6.36 · genuineness 4.62 · WER 0.0%
Fixed-target emotion MSE 0.490 (normalized 0.0306)

1.5 Alpha + real-speech SFT · rank 2

CE 6.40 · genuineness 4.66 · WER 0.0%
Fixed-target emotion MSE 0.595 (normalized 0.0372)

1.5 Alpha + real-speech SFT · rank 4

CE 6.37 · genuineness 4.65 · WER 0.0%
Fixed-target emotion MSE 0.953 (normalized 0.0595)

1.5 Alpha + real-speech SFT · rank 8

CE 6.39 · genuineness 4.65 · WER 0.0%
Fixed-target emotion MSE 0.987 (normalized 0.0617)

fear courage · German · reference

Lass bitte deine Hand auf meiner Schulter; das Licht ist schon wieder aus, und ich kann die Treppe nicht sehen. Hast du das hinter uns gehört? Einen Moment dachte ich... nein, wahrscheinlich war es wirklich nur die Tür. Ich muss kurz durchatmen, weil mein Herz so rast, und ich möchte dir nicht noch mehr Angst machen. Da vorne leuchtet noch dieses kleine grüne Licht; wir gehen gemeinsam an der Wand entlang bis zum Ausgang. Ich habe auch Angst, ganz ehrlich, aber ich bin direkt neben dir, und niemand von uns muss hier alleine durch.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Fear and momentary panic become courage and reassuring companionship.
SCRIPT:
(hushed and alert, genuinely frightened) [6.4 seconds duration] Lass bitte deine Hand auf meiner Schulter; das Licht ist schon wieder aus, und ich kann die Treppe nicht sehen.
[0.3 seconds pause]
(panic briefly breaking through, breath catching) [6.5 seconds duration] Hast du das hinter uns gehört? Einen Moment dachte ich... nein, wahrscheinlich war es wirklich nur die Tür.
[0.3 seconds pause]
(sharp inhale, 0.4 seconds)
(hesitant but trying to steady the breath) [6.8 seconds duration] Ich muss kurz durchatmen, weil mein Herz so rast, und ich möchte dir nicht noch mehr Angst machen.
[0.3 seconds pause]
(gathering courage, clearer and more grounded) [6.8 seconds duration] Da vorne leuchtet noch dieses kleine grüne Licht; wir gehen gemeinsam an der Wand entlang bis zum Ausgang.
[0.3 seconds pause]
(gentle reassurance, still afraid but warmly determined) [6.7 seconds duration] Ich habe auch Angst, ganz ehrlich, aber ich bin direkt neben dir, und niemand von uns muss hier alleine durch.

Fixed seed 20261122 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.36 · genuineness 4.71 · WER 0.0%
Fixed-target emotion MSE 0.973 (normalized 0.0608)

1.5 Alpha · merged, no additional SFT

CE 6.37 · genuineness 4.55 · WER 0.0%
Fixed-target emotion MSE 1.214 (normalized 0.0759)

1.5 Alpha + real-speech SFT · rank 1

CE 6.41 · genuineness 4.49 · WER 0.0%
Fixed-target emotion MSE 1.431 (normalized 0.0895)

1.5 Alpha + real-speech SFT · rank 2

CE 6.37 · genuineness 4.63 · WER 0.0%
Fixed-target emotion MSE 0.273 (normalized 0.0170)

1.5 Alpha + real-speech SFT · rank 4

CE 6.40 · genuineness 4.68 · WER 0.0%
Fixed-target emotion MSE 0.813 (normalized 0.0508)

1.5 Alpha + real-speech SFT · rank 8

CE 6.42 · genuineness 4.68 · WER 0.0%
Fixed-target emotion MSE 0.454 (normalized 0.0284)

fear courage · Dutch · prompt_only

Houd je hand alsjeblieft op mijn schouder; het licht is weer uitgevallen en ik kan de trap niet meer zien. Hoorde jij dat achter ons? Even dacht ik... nee, volgens mij was het toch alleen maar de deur. Ik moet heel even ademhalen, want mijn hart gaat tekeer en ik wil jou niet nog banger maken. Daar brandt nog een klein groen lampje; we kunnen samen langs de muur lopen totdat we bij de uitgang zijn. Ik ben ook bang, echt waar, maar ik blijf naast je en we hoeven dit geen van beiden alleen te doen.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Fear and momentary panic become courage and reassuring companionship.
SCRIPT:
(hushed and alert, genuinely frightened) [6.4 seconds duration] Houd je hand alsjeblieft op mijn schouder; het licht is weer uitgevallen en ik kan de trap niet meer zien.
[0.3 seconds pause]
(panic briefly breaking through, breath catching) [6.5 seconds duration] Hoorde jij dat achter ons? Even dacht ik... nee, volgens mij was het toch alleen maar de deur.
[0.3 seconds pause]
(sharp inhale, 0.4 seconds)
(hesitant but trying to steady the breath) [6.8 seconds duration] Ik moet heel even ademhalen, want mijn hart gaat tekeer en ik wil jou niet nog banger maken.
[0.3 seconds pause]
(gathering courage, clearer and more grounded) [6.8 seconds duration] Daar brandt nog een klein groen lampje; we kunnen samen langs de muur lopen totdat we bij de uitgang zijn.
[0.3 seconds pause]
(gentle reassurance, still afraid but warmly determined) [6.7 seconds duration] Ik ben ook bang, echt waar, maar ik blijf naast je en we hoeven dit geen van beiden alleen te doen.

Fixed seed 20261122 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.28 · genuineness 4.30 · WER 3.1%
Fixed-target emotion MSE 2.762 (normalized 0.1726)

1.5 Alpha · merged, no additional SFT

CE 6.22 · genuineness 4.36 · WER 5.2%
Fixed-target emotion MSE 2.622 (normalized 0.1639)

1.5 Alpha + real-speech SFT · rank 1

CE 6.18 · genuineness 4.33 · WER 3.1%
Fixed-target emotion MSE 2.326 (normalized 0.1454)

1.5 Alpha + real-speech SFT · rank 2

CE 6.28 · genuineness 4.52 · WER 4.1%
Fixed-target emotion MSE 2.902 (normalized 0.1814)

1.5 Alpha + real-speech SFT · rank 4

CE 6.25 · genuineness 4.24 · WER 4.1%
Fixed-target emotion MSE 2.609 (normalized 0.1631)

1.5 Alpha + real-speech SFT · rank 8

CE 6.25 · genuineness 4.43 · WER 4.1%
Fixed-target emotion MSE 1.898 (normalized 0.1186)

fear courage · Dutch · reference

Houd je hand alsjeblieft op mijn schouder; het licht is weer uitgevallen en ik kan de trap niet meer zien. Hoorde jij dat achter ons? Even dacht ik... nee, volgens mij was het toch alleen maar de deur. Ik moet heel even ademhalen, want mijn hart gaat tekeer en ik wil jou niet nog banger maken. Daar brandt nog een klein groen lampje; we kunnen samen langs de muur lopen totdat we bij de uitgang zijn. Ik ben ook bang, echt waar, maar ik blijf naast je en we hoeven dit geen van beiden alleen te doen.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Fear and momentary panic become courage and reassuring companionship.
SCRIPT:
(hushed and alert, genuinely frightened) [6.4 seconds duration] Houd je hand alsjeblieft op mijn schouder; het licht is weer uitgevallen en ik kan de trap niet meer zien.
[0.3 seconds pause]
(panic briefly breaking through, breath catching) [6.5 seconds duration] Hoorde jij dat achter ons? Even dacht ik... nee, volgens mij was het toch alleen maar de deur.
[0.3 seconds pause]
(sharp inhale, 0.4 seconds)
(hesitant but trying to steady the breath) [6.8 seconds duration] Ik moet heel even ademhalen, want mijn hart gaat tekeer en ik wil jou niet nog banger maken.
[0.3 seconds pause]
(gathering courage, clearer and more grounded) [6.8 seconds duration] Daar brandt nog een klein groen lampje; we kunnen samen langs de muur lopen totdat we bij de uitgang zijn.
[0.3 seconds pause]
(gentle reassurance, still afraid but warmly determined) [6.7 seconds duration] Ik ben ook bang, echt waar, maar ik blijf naast je en we hoeven dit geen van beiden alleen te doen.

Fixed seed 20261122 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.24 · genuineness 4.41 · WER 3.1%
Fixed-target emotion MSE 2.702 (normalized 0.1689)

1.5 Alpha · merged, no additional SFT

CE 6.31 · genuineness 4.39 · WER 4.1%
Fixed-target emotion MSE 2.414 (normalized 0.1509)

1.5 Alpha + real-speech SFT · rank 1

CE 6.24 · genuineness 4.46 · WER 3.1%
Fixed-target emotion MSE 2.243 (normalized 0.1402)

1.5 Alpha + real-speech SFT · rank 2

CE 6.34 · genuineness 4.35 · WER 7.2%
Fixed-target emotion MSE 1.946 (normalized 0.1216)

1.5 Alpha + real-speech SFT · rank 4

CE 6.36 · genuineness 4.44 · WER 3.1%
Fixed-target emotion MSE 2.282 (normalized 0.1426)

1.5 Alpha + real-speech SFT · rank 8

CE 6.23 · genuineness 4.33 · WER 3.1%
Fixed-target emotion MSE 2.342 (normalized 0.1464)

fear courage · French · prompt_only

Garde ta main sur mon épaule, s'il te plaît; la lumière s'est encore éteinte et je ne vois plus les marches. Tu as entendu ce bruit derrière nous? Pendant une seconde, j'ai cru... non, c'était probablement juste la porte. Laisse-moi reprendre mon souffle, mon cœur bat trop vite et je ne veux surtout pas te faire encore plus peur. Là-bas, la petite lumière verte fonctionne toujours; nous pouvons longer le mur ensemble jusqu'à la sortie, tout doucement. J'ai peur aussi, vraiment, mais je reste près de toi et personne ne devra traverser ça tout seul.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Fear and momentary panic become courage and reassuring companionship.
SCRIPT:
(hushed and alert, genuinely frightened) [6.4 seconds duration] Garde ta main sur mon épaule, s'il te plaît; la lumière s'est encore éteinte et je ne vois plus les marches.
[0.3 seconds pause]
(panic briefly breaking through, breath catching) [6.5 seconds duration] Tu as entendu ce bruit derrière nous? Pendant une seconde, j'ai cru... non, c'était probablement juste la porte.
[0.3 seconds pause]
(sharp inhale, 0.4 seconds)
(hesitant but trying to steady the breath) [6.8 seconds duration] Laisse-moi reprendre mon souffle, mon cœur bat trop vite et je ne veux surtout pas te faire encore plus peur.
[0.3 seconds pause]
(gathering courage, clearer and more grounded) [6.8 seconds duration] Là-bas, la petite lumière verte fonctionne toujours; nous pouvons longer le mur ensemble jusqu'à la sortie, tout doucement.
[0.3 seconds pause]
(gentle reassurance, still afraid but warmly determined) [6.7 seconds duration] J'ai peur aussi, vraiment, mais je reste près de toi et personne ne devra traverser ça tout seul.

Fixed seed 20261122 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.28 · genuineness 4.76 · WER 0.0%
Fixed-target emotion MSE 2.048 (normalized 0.1280)

1.5 Alpha · merged, no additional SFT

CE 6.21 · genuineness 4.46 · WER 0.0%
Fixed-target emotion MSE 1.760 (normalized 0.1100)

1.5 Alpha + real-speech SFT · rank 1

CE 6.15 · genuineness 4.55 · WER 0.0%
Fixed-target emotion MSE 1.842 (normalized 0.1151)

1.5 Alpha + real-speech SFT · rank 2

CE 6.29 · genuineness 4.51 · WER 0.0%
Fixed-target emotion MSE 2.070 (normalized 0.1294)

1.5 Alpha + real-speech SFT · rank 4

CE 6.17 · genuineness 4.63 · WER 0.0%
Fixed-target emotion MSE 1.572 (normalized 0.0983)

1.5 Alpha + real-speech SFT · rank 8

CE 6.23 · genuineness 4.63 · WER 0.0%
Fixed-target emotion MSE 2.186 (normalized 0.1366)

fear courage · French · reference

Garde ta main sur mon épaule, s'il te plaît; la lumière s'est encore éteinte et je ne vois plus les marches. Tu as entendu ce bruit derrière nous? Pendant une seconde, j'ai cru... non, c'était probablement juste la porte. Laisse-moi reprendre mon souffle, mon cœur bat trop vite et je ne veux surtout pas te faire encore plus peur. Là-bas, la petite lumière verte fonctionne toujours; nous pouvons longer le mur ensemble jusqu'à la sortie, tout doucement. J'ai peur aussi, vraiment, mais je reste près de toi et personne ne devra traverser ça tout seul.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Fear and momentary panic become courage and reassuring companionship.
SCRIPT:
(hushed and alert, genuinely frightened) [6.4 seconds duration] Garde ta main sur mon épaule, s'il te plaît; la lumière s'est encore éteinte et je ne vois plus les marches.
[0.3 seconds pause]
(panic briefly breaking through, breath catching) [6.5 seconds duration] Tu as entendu ce bruit derrière nous? Pendant une seconde, j'ai cru... non, c'était probablement juste la porte.
[0.3 seconds pause]
(sharp inhale, 0.4 seconds)
(hesitant but trying to steady the breath) [6.8 seconds duration] Laisse-moi reprendre mon souffle, mon cœur bat trop vite et je ne veux surtout pas te faire encore plus peur.
[0.3 seconds pause]
(gathering courage, clearer and more grounded) [6.8 seconds duration] Là-bas, la petite lumière verte fonctionne toujours; nous pouvons longer le mur ensemble jusqu'à la sortie, tout doucement.
[0.3 seconds pause]
(gentle reassurance, still afraid but warmly determined) [6.7 seconds duration] J'ai peur aussi, vraiment, mais je reste près de toi et personne ne devra traverser ça tout seul.

Fixed seed 20261122 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.21 · genuineness 4.65 · WER 1.0%
Fixed-target emotion MSE 2.076 (normalized 0.1298)

1.5 Alpha · merged, no additional SFT

CE 5.95 · genuineness 4.43 · WER 3.1%
Fixed-target emotion MSE 2.354 (normalized 0.1471)

1.5 Alpha + real-speech SFT · rank 1

CE 6.07 · genuineness 4.43 · WER 0.0%
Fixed-target emotion MSE 1.758 (normalized 0.1099)

1.5 Alpha + real-speech SFT · rank 2

CE 6.15 · genuineness 4.42 · WER 0.0%
Fixed-target emotion MSE 1.781 (normalized 0.1113)

1.5 Alpha + real-speech SFT · rank 4

CE 6.23 · genuineness 4.51 · WER 0.0%
Fixed-target emotion MSE 1.212 (normalized 0.0758)

1.5 Alpha + real-speech SFT · rank 8

CE 6.20 · genuineness 4.53 · WER 0.0%
Fixed-target emotion MSE 2.361 (normalized 0.1475)

fear courage · Spanish · prompt_only

Deja la mano en mi hombro, por favor; se han vuelto a apagar las luces y no veo los escalones. ¿Has oído eso detrás de nosotros? Por un momento pensé... no, creo que solamente ha sido la puerta. Necesito un segundo para respirar, porque el corazón me va muy rápido y no quiero asustarte todavía más. Mira, esa pequeña luz verde sigue encendida; podemos avanzar juntos siguiendo la pared hasta que lleguemos a la salida. Yo también tengo miedo, de verdad, pero estoy a tu lado y ninguno de los dos va a pasar por esto solo.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Fear and momentary panic become courage and reassuring companionship.
SCRIPT:
(hushed and alert, genuinely frightened) [6.4 seconds duration] Deja la mano en mi hombro, por favor; se han vuelto a apagar las luces y no veo los escalones.
[0.3 seconds pause]
(panic briefly breaking through, breath catching) [6.5 seconds duration] ¿Has oído eso detrás de nosotros? Por un momento pensé... no, creo que solamente ha sido la puerta.
[0.3 seconds pause]
(sharp inhale, 0.4 seconds)
(hesitant but trying to steady the breath) [6.8 seconds duration] Necesito un segundo para respirar, porque el corazón me va muy rápido y no quiero asustarte todavía más.
[0.3 seconds pause]
(gathering courage, clearer and more grounded) [6.8 seconds duration] Mira, esa pequeña luz verde sigue encendida; podemos avanzar juntos siguiendo la pared hasta que lleguemos a la salida.
[0.3 seconds pause]
(gentle reassurance, still afraid but warmly determined) [6.7 seconds duration] Yo también tengo miedo, de verdad, pero estoy a tu lado y ninguno de los dos va a pasar por esto solo.

Fixed seed 20261122 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.17 · genuineness 4.54 · WER 1.0%
Fixed-target emotion MSE 1.944 (normalized 0.1215)

1.5 Alpha · merged, no additional SFT

CE 6.23 · genuineness 4.49 · WER 0.0%
Fixed-target emotion MSE 1.537 (normalized 0.0960)

1.5 Alpha + real-speech SFT · rank 1

CE 5.99 · genuineness 4.54 · WER 0.0%
Fixed-target emotion MSE 1.556 (normalized 0.0973)

1.5 Alpha + real-speech SFT · rank 2

CE 6.20 · genuineness 4.41 · WER 0.0%
Fixed-target emotion MSE 1.420 (normalized 0.0888)

1.5 Alpha + real-speech SFT · rank 4

CE 6.27 · genuineness 4.54 · WER 0.0%
Fixed-target emotion MSE 0.999 (normalized 0.0624)

1.5 Alpha + real-speech SFT · rank 8

CE 6.22 · genuineness 4.53 · WER 0.0%
Fixed-target emotion MSE 1.963 (normalized 0.1227)

fear courage · Spanish · reference

Deja la mano en mi hombro, por favor; se han vuelto a apagar las luces y no veo los escalones. ¿Has oído eso detrás de nosotros? Por un momento pensé... no, creo que solamente ha sido la puerta. Necesito un segundo para respirar, porque el corazón me va muy rápido y no quiero asustarte todavía más. Mira, esa pequeña luz verde sigue encendida; podemos avanzar juntos siguiendo la pared hasta que lleguemos a la salida. Yo también tengo miedo, de verdad, pero estoy a tu lado y ninguno de los dos va a pasar por esto solo.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Fear and momentary panic become courage and reassuring companionship.
SCRIPT:
(hushed and alert, genuinely frightened) [6.4 seconds duration] Deja la mano en mi hombro, por favor; se han vuelto a apagar las luces y no veo los escalones.
[0.3 seconds pause]
(panic briefly breaking through, breath catching) [6.5 seconds duration] ¿Has oído eso detrás de nosotros? Por un momento pensé... no, creo que solamente ha sido la puerta.
[0.3 seconds pause]
(sharp inhale, 0.4 seconds)
(hesitant but trying to steady the breath) [6.8 seconds duration] Necesito un segundo para respirar, porque el corazón me va muy rápido y no quiero asustarte todavía más.
[0.3 seconds pause]
(gathering courage, clearer and more grounded) [6.8 seconds duration] Mira, esa pequeña luz verde sigue encendida; podemos avanzar juntos siguiendo la pared hasta que lleguemos a la salida.
[0.3 seconds pause]
(gentle reassurance, still afraid but warmly determined) [6.7 seconds duration] Yo también tengo miedo, de verdad, pero estoy a tu lado y ninguno de los dos va a pasar por esto solo.

Fixed seed 20261122 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.28 · genuineness 4.32 · WER 0.0%
Fixed-target emotion MSE 2.554 (normalized 0.1596)

1.5 Alpha · merged, no additional SFT

CE 6.16 · genuineness 4.42 · WER 0.0%
Fixed-target emotion MSE 2.489 (normalized 0.1555)

1.5 Alpha + real-speech SFT · rank 1

CE 6.22 · genuineness 4.42 · WER 0.0%
Fixed-target emotion MSE 2.303 (normalized 0.1439)

1.5 Alpha + real-speech SFT · rank 2

CE 6.22 · genuineness 4.49 · WER 0.0%
Fixed-target emotion MSE 2.325 (normalized 0.1453)

1.5 Alpha + real-speech SFT · rank 4

CE 6.28 · genuineness 4.36 · WER 0.0%
Fixed-target emotion MSE 2.800 (normalized 0.1750)

1.5 Alpha + real-speech SFT · rank 8

CE 6.24 · genuineness 4.37 · WER 0.0%
Fixed-target emotion MSE 2.755 (normalized 0.1722)

grief gratitude · English · prompt_only

I nearly threw this old radio away last winter, because every time I touched it I remembered his hands. He used to turn that crooked little dial and insist the music sounded better when the kitchen window was open. And now you have fixed it... I did not realize how much I wanted to hear that silly crackle again. Thank you for keeping the scratches instead of making it look new; they are part of what I missed. Could you stay for a cup of tea while it plays? I think I would like to remember him with someone tonight.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Grief, a fond memory and unexpectedly tender gratitude.
SCRIPT:
(quiet grief, slightly trembling but intelligible) [6.4 seconds duration] I nearly threw this old radio away last winter, because every time I touched it I remembered his hands.
[0.3 seconds pause]
(a fond memory bringing a fragile smile) [6.5 seconds duration] He used to turn that crooked little dial and insist the music sounded better when the kitchen window was open.
[0.3 seconds pause]
(wistful sigh, 0.4 seconds)
(voice catching, allow a small reflective hesitation) [6.8 seconds duration] And now you have fixed it... I did not realize how much I wanted to hear that silly crackle again.
[0.3 seconds pause]
(surprised gratitude, intimate and sincere) [6.8 seconds duration] Thank you for keeping the scratches instead of making it look new; they are part of what I missed.
[0.3 seconds pause]
(steadier, warm and quietly hopeful) [6.7 seconds duration] Could you stay for a cup of tea while it plays? I think I would like to remember him with someone tonight.

Fixed seed 20261223 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.61 · genuineness 4.33 · WER 0.0%
Fixed-target emotion MSE 1.549 (normalized 0.0968)

1.5 Alpha · merged, no additional SFT

CE 6.54 · genuineness 4.25 · WER 1.0%
Fixed-target emotion MSE 1.552 (normalized 0.0970)

1.5 Alpha + real-speech SFT · rank 1

CE 6.53 · genuineness 4.20 · WER 0.0%
Fixed-target emotion MSE 1.568 (normalized 0.0980)

1.5 Alpha + real-speech SFT · rank 2

CE 6.51 · genuineness 4.21 · WER 1.0%
Fixed-target emotion MSE 1.441 (normalized 0.0901)

1.5 Alpha + real-speech SFT · rank 4

CE 6.57 · genuineness 4.14 · WER 0.0%
Fixed-target emotion MSE 1.518 (normalized 0.0949)

1.5 Alpha + real-speech SFT · rank 8

CE 6.53 · genuineness 4.28 · WER 0.0%
Fixed-target emotion MSE 1.626 (normalized 0.1016)

grief gratitude · English · reference

I nearly threw this old radio away last winter, because every time I touched it I remembered his hands. He used to turn that crooked little dial and insist the music sounded better when the kitchen window was open. And now you have fixed it... I did not realize how much I wanted to hear that silly crackle again. Thank you for keeping the scratches instead of making it look new; they are part of what I missed. Could you stay for a cup of tea while it plays? I think I would like to remember him with someone tonight.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Grief, a fond memory and unexpectedly tender gratitude.
SCRIPT:
(quiet grief, slightly trembling but intelligible) [6.4 seconds duration] I nearly threw this old radio away last winter, because every time I touched it I remembered his hands.
[0.3 seconds pause]
(a fond memory bringing a fragile smile) [6.5 seconds duration] He used to turn that crooked little dial and insist the music sounded better when the kitchen window was open.
[0.3 seconds pause]
(wistful sigh, 0.4 seconds)
(voice catching, allow a small reflective hesitation) [6.8 seconds duration] And now you have fixed it... I did not realize how much I wanted to hear that silly crackle again.
[0.3 seconds pause]
(surprised gratitude, intimate and sincere) [6.8 seconds duration] Thank you for keeping the scratches instead of making it look new; they are part of what I missed.
[0.3 seconds pause]
(steadier, warm and quietly hopeful) [6.7 seconds duration] Could you stay for a cup of tea while it plays? I think I would like to remember him with someone tonight.

Fixed seed 20261223 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.56 · genuineness 4.15 · WER 0.0%
Fixed-target emotion MSE 1.558 (normalized 0.0974)

1.5 Alpha · merged, no additional SFT

CE 6.58 · genuineness 4.11 · WER 0.0%
Fixed-target emotion MSE 1.732 (normalized 0.1082)

1.5 Alpha + real-speech SFT · rank 1

CE 6.52 · genuineness 4.31 · WER 0.0%
Fixed-target emotion MSE 1.451 (normalized 0.0907)

1.5 Alpha + real-speech SFT · rank 2

CE 6.52 · genuineness 4.17 · WER 0.0%
Fixed-target emotion MSE 1.499 (normalized 0.0937)

1.5 Alpha + real-speech SFT · rank 4

CE 6.58 · genuineness 4.22 · WER 0.0%
Fixed-target emotion MSE 1.622 (normalized 0.1013)

1.5 Alpha + real-speech SFT · rank 8

CE 6.60 · genuineness 4.14 · WER 0.0%
Fixed-target emotion MSE 1.738 (normalized 0.1086)

grief gratitude · German · prompt_only

Letzten Winter wollte ich dieses alte Radio fast wegwerfen, weil ich jedes Mal seine Hände vor mir gesehen habe. Er drehte immer an diesem schiefen kleinen Knopf und behauptete, bei offenem Küchenfenster würde die Musik einfach besser klingen. Und jetzt hast du es repariert... mir war gar nicht klar, wie sehr ich dieses alberne Knistern vermisst habe. Danke, dass du die Kratzer nicht wegpoliert hast; genau diese kleinen Spuren gehören zu dem, was mir gefehlt hat. Bleibst du noch auf einen Tee, während es spielt? Ich glaube, heute möchte ich mich nicht ganz alleine an ihn erinnern.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Grief, a fond memory and unexpectedly tender gratitude.
SCRIPT:
(quiet grief, slightly trembling but intelligible) [6.4 seconds duration] Letzten Winter wollte ich dieses alte Radio fast wegwerfen, weil ich jedes Mal seine Hände vor mir gesehen habe.
[0.3 seconds pause]
(a fond memory bringing a fragile smile) [6.5 seconds duration] Er drehte immer an diesem schiefen kleinen Knopf und behauptete, bei offenem Küchenfenster würde die Musik einfach besser klingen.
[0.3 seconds pause]
(wistful sigh, 0.4 seconds)
(voice catching, allow a small reflective hesitation) [6.8 seconds duration] Und jetzt hast du es repariert... mir war gar nicht klar, wie sehr ich dieses alberne Knistern vermisst habe.
[0.3 seconds pause]
(surprised gratitude, intimate and sincere) [6.8 seconds duration] Danke, dass du die Kratzer nicht wegpoliert hast; genau diese kleinen Spuren gehören zu dem, was mir gefehlt hat.
[0.3 seconds pause]
(steadier, warm and quietly hopeful) [6.7 seconds duration] Bleibst du noch auf einen Tee, während es spielt? Ich glaube, heute möchte ich mich nicht ganz alleine an ihn erinnern.

Fixed seed 20261223 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.33 · genuineness 4.51 · WER 0.0%
Fixed-target emotion MSE 1.819 (normalized 0.1137)

1.5 Alpha · merged, no additional SFT

CE 6.38 · genuineness 4.38 · WER 0.0%
Fixed-target emotion MSE 1.647 (normalized 0.1029)

1.5 Alpha + real-speech SFT · rank 1

CE 6.25 · genuineness 4.28 · WER 0.0%
Fixed-target emotion MSE 1.381 (normalized 0.0863)

1.5 Alpha + real-speech SFT · rank 2

CE 6.42 · genuineness 4.47 · WER 0.0%
Fixed-target emotion MSE 1.387 (normalized 0.0867)

1.5 Alpha + real-speech SFT · rank 4

CE 6.37 · genuineness 4.48 · WER 0.0%
Fixed-target emotion MSE 1.156 (normalized 0.0722)

1.5 Alpha + real-speech SFT · rank 8

CE 6.36 · genuineness 4.29 · WER 0.0%
Fixed-target emotion MSE 1.089 (normalized 0.0681)

grief gratitude · German · reference

Letzten Winter wollte ich dieses alte Radio fast wegwerfen, weil ich jedes Mal seine Hände vor mir gesehen habe. Er drehte immer an diesem schiefen kleinen Knopf und behauptete, bei offenem Küchenfenster würde die Musik einfach besser klingen. Und jetzt hast du es repariert... mir war gar nicht klar, wie sehr ich dieses alberne Knistern vermisst habe. Danke, dass du die Kratzer nicht wegpoliert hast; genau diese kleinen Spuren gehören zu dem, was mir gefehlt hat. Bleibst du noch auf einen Tee, während es spielt? Ich glaube, heute möchte ich mich nicht ganz alleine an ihn erinnern.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Grief, a fond memory and unexpectedly tender gratitude.
SCRIPT:
(quiet grief, slightly trembling but intelligible) [6.4 seconds duration] Letzten Winter wollte ich dieses alte Radio fast wegwerfen, weil ich jedes Mal seine Hände vor mir gesehen habe.
[0.3 seconds pause]
(a fond memory bringing a fragile smile) [6.5 seconds duration] Er drehte immer an diesem schiefen kleinen Knopf und behauptete, bei offenem Küchenfenster würde die Musik einfach besser klingen.
[0.3 seconds pause]
(wistful sigh, 0.4 seconds)
(voice catching, allow a small reflective hesitation) [6.8 seconds duration] Und jetzt hast du es repariert... mir war gar nicht klar, wie sehr ich dieses alberne Knistern vermisst habe.
[0.3 seconds pause]
(surprised gratitude, intimate and sincere) [6.8 seconds duration] Danke, dass du die Kratzer nicht wegpoliert hast; genau diese kleinen Spuren gehören zu dem, was mir gefehlt hat.
[0.3 seconds pause]
(steadier, warm and quietly hopeful) [6.7 seconds duration] Bleibst du noch auf einen Tee, während es spielt? Ich glaube, heute möchte ich mich nicht ganz alleine an ihn erinnern.

Fixed seed 20261223 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.40 · genuineness 4.50 · WER 0.0%
Fixed-target emotion MSE 1.308 (normalized 0.0817)

1.5 Alpha · merged, no additional SFT

CE 6.48 · genuineness 4.43 · WER 1.0%
Fixed-target emotion MSE 2.397 (normalized 0.1498)

1.5 Alpha + real-speech SFT · rank 1

CE 6.40 · genuineness 4.53 · WER 0.0%
Fixed-target emotion MSE 1.017 (normalized 0.0636)

1.5 Alpha + real-speech SFT · rank 2

CE 6.29 · genuineness 4.46 · WER 1.0%
Fixed-target emotion MSE 1.518 (normalized 0.0949)

1.5 Alpha + real-speech SFT · rank 4

CE 6.44 · genuineness 4.39 · WER 0.0%
Fixed-target emotion MSE 1.340 (normalized 0.0837)

1.5 Alpha + real-speech SFT · rank 8

CE 6.43 · genuineness 4.58 · WER 1.0%
Fixed-target emotion MSE 1.497 (normalized 0.0936)

grief gratitude · Dutch · prompt_only

Vorige winter had ik deze oude radio bijna weggegooid, omdat ik bij elke aanraking weer aan zijn handen moest denken. Hij draaide altijd aan dat scheve knopje en beweerde dat de muziek beter klonk als het keukenraam openstond. En nu heb jij hem gerepareerd... ik wist niet eens hoe graag ik dat gekke gekraak weer wilde horen. Dank je dat je de krassen hebt laten zitten in plaats van alles nieuw te maken; die kleine sporen horen erbij. Wil je nog een kop thee blijven drinken terwijl hij speelt? Ik zou vanavond graag samen met iemand aan hem denken.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Grief, a fond memory and unexpectedly tender gratitude.
SCRIPT:
(quiet grief, slightly trembling but intelligible) [6.4 seconds duration] Vorige winter had ik deze oude radio bijna weggegooid, omdat ik bij elke aanraking weer aan zijn handen moest denken.
[0.3 seconds pause]
(a fond memory bringing a fragile smile) [6.5 seconds duration] Hij draaide altijd aan dat scheve knopje en beweerde dat de muziek beter klonk als het keukenraam openstond.
[0.3 seconds pause]
(wistful sigh, 0.4 seconds)
(voice catching, allow a small reflective hesitation) [6.8 seconds duration] En nu heb jij hem gerepareerd... ik wist niet eens hoe graag ik dat gekke gekraak weer wilde horen.
[0.3 seconds pause]
(surprised gratitude, intimate and sincere) [6.8 seconds duration] Dank je dat je de krassen hebt laten zitten in plaats van alles nieuw te maken; die kleine sporen horen erbij.
[0.3 seconds pause]
(steadier, warm and quietly hopeful) [6.7 seconds duration] Wil je nog een kop thee blijven drinken terwijl hij speelt? Ik zou vanavond graag samen met iemand aan hem denken.

Fixed seed 20261223 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.30 · genuineness 4.43 · WER 4.0%
Fixed-target emotion MSE 3.291 (normalized 0.2057)

1.5 Alpha · merged, no additional SFT

CE 6.30 · genuineness 4.41 · WER 6.1%
Fixed-target emotion MSE 3.480 (normalized 0.2175)

1.5 Alpha + real-speech SFT · rank 1

CE 6.27 · genuineness 4.40 · WER 5.1%
Fixed-target emotion MSE 3.198 (normalized 0.1999)

1.5 Alpha + real-speech SFT · rank 2

CE 6.33 · genuineness 4.39 · WER 7.1%
Fixed-target emotion MSE 3.327 (normalized 0.2079)

1.5 Alpha + real-speech SFT · rank 4

CE 6.27 · genuineness 4.43 · WER 5.1%
Fixed-target emotion MSE 2.971 (normalized 0.1857)

1.5 Alpha + real-speech SFT · rank 8

CE 6.26 · genuineness 4.34 · WER 6.1%
Fixed-target emotion MSE 3.315 (normalized 0.2072)

grief gratitude · Dutch · reference

Vorige winter had ik deze oude radio bijna weggegooid, omdat ik bij elke aanraking weer aan zijn handen moest denken. Hij draaide altijd aan dat scheve knopje en beweerde dat de muziek beter klonk als het keukenraam openstond. En nu heb jij hem gerepareerd... ik wist niet eens hoe graag ik dat gekke gekraak weer wilde horen. Dank je dat je de krassen hebt laten zitten in plaats van alles nieuw te maken; die kleine sporen horen erbij. Wil je nog een kop thee blijven drinken terwijl hij speelt? Ik zou vanavond graag samen met iemand aan hem denken.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Grief, a fond memory and unexpectedly tender gratitude.
SCRIPT:
(quiet grief, slightly trembling but intelligible) [6.4 seconds duration] Vorige winter had ik deze oude radio bijna weggegooid, omdat ik bij elke aanraking weer aan zijn handen moest denken.
[0.3 seconds pause]
(a fond memory bringing a fragile smile) [6.5 seconds duration] Hij draaide altijd aan dat scheve knopje en beweerde dat de muziek beter klonk als het keukenraam openstond.
[0.3 seconds pause]
(wistful sigh, 0.4 seconds)
(voice catching, allow a small reflective hesitation) [6.8 seconds duration] En nu heb jij hem gerepareerd... ik wist niet eens hoe graag ik dat gekke gekraak weer wilde horen.
[0.3 seconds pause]
(surprised gratitude, intimate and sincere) [6.8 seconds duration] Dank je dat je de krassen hebt laten zitten in plaats van alles nieuw te maken; die kleine sporen horen erbij.
[0.3 seconds pause]
(steadier, warm and quietly hopeful) [6.7 seconds duration] Wil je nog een kop thee blijven drinken terwijl hij speelt? Ik zou vanavond graag samen met iemand aan hem denken.

Fixed seed 20261223 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.29 · genuineness 4.33 · WER 5.1%
Fixed-target emotion MSE 3.334 (normalized 0.2084)

1.5 Alpha · merged, no additional SFT

CE 6.32 · genuineness 4.34 · WER 6.1%
Fixed-target emotion MSE 2.994 (normalized 0.1872)

1.5 Alpha + real-speech SFT · rank 1

CE 6.23 · genuineness 4.14 · WER 6.1%
Fixed-target emotion MSE 2.902 (normalized 0.1814)

1.5 Alpha + real-speech SFT · rank 2

CE 6.27 · genuineness 4.10 · WER 6.1%
Fixed-target emotion MSE 3.399 (normalized 0.2124)

1.5 Alpha + real-speech SFT · rank 4

CE 6.25 · genuineness 4.01 · WER 5.1%
Fixed-target emotion MSE 3.532 (normalized 0.2208)

1.5 Alpha + real-speech SFT · rank 8

CE 6.29 · genuineness 4.17 · WER 6.1%
Fixed-target emotion MSE 2.835 (normalized 0.1772)

grief gratitude · French · prompt_only

L'hiver dernier, j'ai failli jeter cette vieille radio, parce que chaque fois que je la touchais, je revoyais ses mains. Il tournait toujours ce petit bouton de travers et disait que la musique était meilleure quand la fenêtre restait ouverte. Et maintenant, tu l'as réparée... je ne savais pas à quel point ce petit grésillement ridicule me manquait encore. Merci d'avoir gardé les rayures au lieu de tout remettre à neuf; ces petites marques font partie du souvenir. Tu resterais prendre un thé pendant qu'elle joue? Je crois que ce soir, je voudrais penser à lui avec quelqu'un.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Grief, a fond memory and unexpectedly tender gratitude.
SCRIPT:
(quiet grief, slightly trembling but intelligible) [6.4 seconds duration] L'hiver dernier, j'ai failli jeter cette vieille radio, parce que chaque fois que je la touchais, je revoyais ses mains.
[0.3 seconds pause]
(a fond memory bringing a fragile smile) [6.5 seconds duration] Il tournait toujours ce petit bouton de travers et disait que la musique était meilleure quand la fenêtre restait ouverte.
[0.3 seconds pause]
(wistful sigh, 0.4 seconds)
(voice catching, allow a small reflective hesitation) [6.8 seconds duration] Et maintenant, tu l'as réparée... je ne savais pas à quel point ce petit grésillement ridicule me manquait encore.
[0.3 seconds pause]
(surprised gratitude, intimate and sincere) [6.8 seconds duration] Merci d'avoir gardé les rayures au lieu de tout remettre à neuf; ces petites marques font partie du souvenir.
[0.3 seconds pause]
(steadier, warm and quietly hopeful) [6.7 seconds duration] Tu resterais prendre un thé pendant qu'elle joue? Je crois que ce soir, je voudrais penser à lui avec quelqu'un.

Fixed seed 20261223 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.20 · genuineness 4.49 · WER 13.3%
Fixed-target emotion MSE 1.032 (normalized 0.0645)

1.5 Alpha · merged, no additional SFT

CE 6.23 · genuineness 4.49 · WER 19.4%
Fixed-target emotion MSE 1.138 (normalized 0.0711)

1.5 Alpha + real-speech SFT · rank 1

CE 6.21 · genuineness 4.23 · WER 2.0%
Fixed-target emotion MSE 1.459 (normalized 0.0912)

1.5 Alpha + real-speech SFT · rank 2

CE 6.13 · genuineness 4.32 · WER 2.0%
Fixed-target emotion MSE 1.352 (normalized 0.0845)

1.5 Alpha + real-speech SFT · rank 4

CE 6.10 · genuineness 4.37 · WER 1.0%
Fixed-target emotion MSE 1.422 (normalized 0.0889)

1.5 Alpha + real-speech SFT · rank 8

CE 6.17 · genuineness 4.39 · WER 1.0%
Fixed-target emotion MSE 1.513 (normalized 0.0946)

grief gratitude · French · reference

L'hiver dernier, j'ai failli jeter cette vieille radio, parce que chaque fois que je la touchais, je revoyais ses mains. Il tournait toujours ce petit bouton de travers et disait que la musique était meilleure quand la fenêtre restait ouverte. Et maintenant, tu l'as réparée... je ne savais pas à quel point ce petit grésillement ridicule me manquait encore. Merci d'avoir gardé les rayures au lieu de tout remettre à neuf; ces petites marques font partie du souvenir. Tu resterais prendre un thé pendant qu'elle joue? Je crois que ce soir, je voudrais penser à lui avec quelqu'un.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Grief, a fond memory and unexpectedly tender gratitude.
SCRIPT:
(quiet grief, slightly trembling but intelligible) [6.4 seconds duration] L'hiver dernier, j'ai failli jeter cette vieille radio, parce que chaque fois que je la touchais, je revoyais ses mains.
[0.3 seconds pause]
(a fond memory bringing a fragile smile) [6.5 seconds duration] Il tournait toujours ce petit bouton de travers et disait que la musique était meilleure quand la fenêtre restait ouverte.
[0.3 seconds pause]
(wistful sigh, 0.4 seconds)
(voice catching, allow a small reflective hesitation) [6.8 seconds duration] Et maintenant, tu l'as réparée... je ne savais pas à quel point ce petit grésillement ridicule me manquait encore.
[0.3 seconds pause]
(surprised gratitude, intimate and sincere) [6.8 seconds duration] Merci d'avoir gardé les rayures au lieu de tout remettre à neuf; ces petites marques font partie du souvenir.
[0.3 seconds pause]
(steadier, warm and quietly hopeful) [6.7 seconds duration] Tu resterais prendre un thé pendant qu'elle joue? Je crois que ce soir, je voudrais penser à lui avec quelqu'un.

Fixed seed 20261223 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 5.98 · genuineness 4.54 · WER 2.0%
Fixed-target emotion MSE 1.297 (normalized 0.0811)

1.5 Alpha · merged, no additional SFT

CE 6.16 · genuineness 4.51 · WER 3.1%
Fixed-target emotion MSE 1.280 (normalized 0.0800)

1.5 Alpha + real-speech SFT · rank 1

CE 6.08 · genuineness 4.63 · WER 1.0%
Fixed-target emotion MSE 0.667 (normalized 0.0417)

1.5 Alpha + real-speech SFT · rank 2

CE 6.11 · genuineness 4.46 · WER 3.1%
Fixed-target emotion MSE 1.404 (normalized 0.0878)

1.5 Alpha + real-speech SFT · rank 4

CE 6.12 · genuineness 4.56 · WER 1.0%
Fixed-target emotion MSE 0.779 (normalized 0.0487)

1.5 Alpha + real-speech SFT · rank 8

CE 6.18 · genuineness 4.31 · WER 1.0%
Fixed-target emotion MSE 1.485 (normalized 0.0928)

grief gratitude · Spanish · prompt_only

El invierno pasado casi tiré esta vieja radio, porque cada vez que la tocaba volvía a recordar sus manos. Siempre giraba ese pequeño botón torcido y decía que la música sonaba mejor con la ventana de la cocina abierta. Y ahora la has arreglado... no sabía cuánto necesitaba volver a escuchar ese pequeño crujido que antes me parecía absurdo. Gracias por conservar los arañazos en vez de dejarla como nueva; esas marcas también forman parte de lo que echaba de menos. ¿Te quedarías a tomar un té mientras suena? Creo que esta noche me gustaría recordarlo en compañía de alguien.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Grief, a fond memory and unexpectedly tender gratitude.
SCRIPT:
(quiet grief, slightly trembling but intelligible) [6.4 seconds duration] El invierno pasado casi tiré esta vieja radio, porque cada vez que la tocaba volvía a recordar sus manos.
[0.3 seconds pause]
(a fond memory bringing a fragile smile) [6.5 seconds duration] Siempre giraba ese pequeño botón torcido y decía que la música sonaba mejor con la ventana de la cocina abierta.
[0.3 seconds pause]
(wistful sigh, 0.4 seconds)
(voice catching, allow a small reflective hesitation) [6.8 seconds duration] Y ahora la has arreglado... no sabía cuánto necesitaba volver a escuchar ese pequeño crujido que antes me parecía absurdo.
[0.3 seconds pause]
(surprised gratitude, intimate and sincere) [6.8 seconds duration] Gracias por conservar los arañazos en vez de dejarla como nueva; esas marcas también forman parte de lo que echaba de menos.
[0.3 seconds pause]
(steadier, warm and quietly hopeful) [6.7 seconds duration] ¿Te quedarías a tomar un té mientras suena? Creo que esta noche me gustaría recordarlo en compañía de alguien.

Fixed seed 20261223 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.21 · genuineness 4.29 · WER 1.0%
Fixed-target emotion MSE 1.569 (normalized 0.0981)

1.5 Alpha · merged, no additional SFT

CE 6.28 · genuineness 4.42 · WER 1.0%
Fixed-target emotion MSE 1.364 (normalized 0.0853)

1.5 Alpha + real-speech SFT · rank 1

CE 6.23 · genuineness 4.34 · WER 1.0%
Fixed-target emotion MSE 1.366 (normalized 0.0854)

1.5 Alpha + real-speech SFT · rank 2

CE 6.17 · genuineness 4.19 · WER 1.0%
Fixed-target emotion MSE 1.538 (normalized 0.0961)

1.5 Alpha + real-speech SFT · rank 4

CE 6.31 · genuineness 4.48 · WER 1.0%
Fixed-target emotion MSE 1.351 (normalized 0.0844)

1.5 Alpha + real-speech SFT · rank 8

CE 6.13 · genuineness 4.29 · WER 1.0%
Fixed-target emotion MSE 1.081 (normalized 0.0675)

grief gratitude · Spanish · reference

El invierno pasado casi tiré esta vieja radio, porque cada vez que la tocaba volvía a recordar sus manos. Siempre giraba ese pequeño botón torcido y decía que la música sonaba mejor con la ventana de la cocina abierta. Y ahora la has arreglado... no sabía cuánto necesitaba volver a escuchar ese pequeño crujido que antes me parecía absurdo. Gracias por conservar los arañazos en vez de dejarla como nueva; esas marcas también forman parte de lo que echaba de menos. ¿Te quedarías a tomar un té mientras suena? Creo que esta noche me gustaría recordarlo en compañía de alguien.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Grief, a fond memory and unexpectedly tender gratitude.
SCRIPT:
(quiet grief, slightly trembling but intelligible) [6.4 seconds duration] El invierno pasado casi tiré esta vieja radio, porque cada vez que la tocaba volvía a recordar sus manos.
[0.3 seconds pause]
(a fond memory bringing a fragile smile) [6.5 seconds duration] Siempre giraba ese pequeño botón torcido y decía que la música sonaba mejor con la ventana de la cocina abierta.
[0.3 seconds pause]
(wistful sigh, 0.4 seconds)
(voice catching, allow a small reflective hesitation) [6.8 seconds duration] Y ahora la has arreglado... no sabía cuánto necesitaba volver a escuchar ese pequeño crujido que antes me parecía absurdo.
[0.3 seconds pause]
(surprised gratitude, intimate and sincere) [6.8 seconds duration] Gracias por conservar los arañazos en vez de dejarla como nueva; esas marcas también forman parte de lo que echaba de menos.
[0.3 seconds pause]
(steadier, warm and quietly hopeful) [6.7 seconds duration] ¿Te quedarías a tomar un té mientras suena? Creo que esta noche me gustaría recordarlo en compañía de alguien.

Fixed seed 20261223 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.20 · genuineness 4.23 · WER 1.0%
Fixed-target emotion MSE 1.615 (normalized 0.1009)

1.5 Alpha · merged, no additional SFT

CE 6.10 · genuineness 4.16 · WER 1.0%
Fixed-target emotion MSE 1.165 (normalized 0.0728)

1.5 Alpha + real-speech SFT · rank 1

CE 6.09 · genuineness 4.17 · WER 1.0%
Fixed-target emotion MSE 1.508 (normalized 0.0942)

1.5 Alpha + real-speech SFT · rank 2

CE 6.26 · genuineness 4.25 · WER 1.0%
Fixed-target emotion MSE 1.571 (normalized 0.0982)

1.5 Alpha + real-speech SFT · rank 4

CE 6.07 · genuineness 4.22 · WER 1.0%
Fixed-target emotion MSE 1.965 (normalized 0.1228)

1.5 Alpha + real-speech SFT · rank 8

CE 6.25 · genuineness 4.11 · WER 3.0%
Fixed-target emotion MSE 1.554 (normalized 0.0971)

joy disbelief · English · prompt_only

Wait, do not say anything yet; I have read the first line three times and I still cannot believe it. They are offering me the place, the actual place, with the little workshop overlooking the sea that we talked about. I got it... after all those applications and all those terrible interviews, somebody finally wants me to come. You stayed up with me through every doubtful evening, and I need you to know that I remember all of it. Come here, please, let me hug you before I call everyone; I want this first happy moment to be ours.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Disbelief blossoms into exuberant joy and heartfelt gratitude.
SCRIPT:
(very quiet disbelief, rereading the news aloud) [6.4 seconds duration] Wait, do not say anything yet; I have read the first line three times and I still cannot believe it.
[0.3 seconds pause]
(hope rising, slightly breathless with excitement) [6.5 seconds duration] They are offering me the place, the actual place, with the little workshop overlooking the sea that we talked about.
[0.3 seconds pause]
(breathy giggle, 0.4 seconds)
(exuberantly delighted, a spontaneous bright laugh nearby) [6.8 seconds duration] I got it... after all those applications and all those terrible interviews, somebody finally wants me to come.
[0.3 seconds pause]
(overwhelming gratitude, let the smile soften) [6.8 seconds duration] You stayed up with me through every doubtful evening, and I need you to know that I remember all of it.
[0.3 seconds pause]
(warmly elated, sharing rather than announcing) [6.7 seconds duration] Come here, please, let me hug you before I call everyone; I want this first happy moment to be ours.

Fixed seed 20261324 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.39 · genuineness 4.06 · WER 0.0%
Fixed-target emotion MSE 1.323 (normalized 0.0827)

1.5 Alpha · merged, no additional SFT

CE 6.50 · genuineness 4.03 · WER 1.0%
Fixed-target emotion MSE 1.121 (normalized 0.0701)

1.5 Alpha + real-speech SFT · rank 1

CE 6.49 · genuineness 4.26 · WER 0.0%
Fixed-target emotion MSE 1.132 (normalized 0.0707)

1.5 Alpha + real-speech SFT · rank 2

CE 6.51 · genuineness 4.02 · WER 0.0%
Fixed-target emotion MSE 1.174 (normalized 0.0734)

1.5 Alpha + real-speech SFT · rank 4

CE 6.46 · genuineness 4.02 · WER 0.0%
Fixed-target emotion MSE 0.707 (normalized 0.0442)

1.5 Alpha + real-speech SFT · rank 8

CE 6.35 · genuineness 3.88 · WER 0.0%
Fixed-target emotion MSE 1.379 (normalized 0.0862)

joy disbelief · English · reference

Wait, do not say anything yet; I have read the first line three times and I still cannot believe it. They are offering me the place, the actual place, with the little workshop overlooking the sea that we talked about. I got it... after all those applications and all those terrible interviews, somebody finally wants me to come. You stayed up with me through every doubtful evening, and I need you to know that I remember all of it. Come here, please, let me hug you before I call everyone; I want this first happy moment to be ours.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Disbelief blossoms into exuberant joy and heartfelt gratitude.
SCRIPT:
(very quiet disbelief, rereading the news aloud) [6.4 seconds duration] Wait, do not say anything yet; I have read the first line three times and I still cannot believe it.
[0.3 seconds pause]
(hope rising, slightly breathless with excitement) [6.5 seconds duration] They are offering me the place, the actual place, with the little workshop overlooking the sea that we talked about.
[0.3 seconds pause]
(breathy giggle, 0.4 seconds)
(exuberantly delighted, a spontaneous bright laugh nearby) [6.8 seconds duration] I got it... after all those applications and all those terrible interviews, somebody finally wants me to come.
[0.3 seconds pause]
(overwhelming gratitude, let the smile soften) [6.8 seconds duration] You stayed up with me through every doubtful evening, and I need you to know that I remember all of it.
[0.3 seconds pause]
(warmly elated, sharing rather than announcing) [6.7 seconds duration] Come here, please, let me hug you before I call everyone; I want this first happy moment to be ours.

Fixed seed 20261324 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.52 · genuineness 4.01 · WER 0.0%
Fixed-target emotion MSE 1.018 (normalized 0.0637)

1.5 Alpha · merged, no additional SFT

CE 6.47 · genuineness 4.03 · WER 0.0%
Fixed-target emotion MSE 1.142 (normalized 0.0714)

1.5 Alpha + real-speech SFT · rank 1

CE 6.49 · genuineness 4.19 · WER 0.0%
Fixed-target emotion MSE 1.405 (normalized 0.0878)

1.5 Alpha + real-speech SFT · rank 2

CE 6.53 · genuineness 4.12 · WER 0.0%
Fixed-target emotion MSE 1.313 (normalized 0.0820)

1.5 Alpha + real-speech SFT · rank 4

CE 6.45 · genuineness 3.99 · WER 0.0%
Fixed-target emotion MSE 0.816 (normalized 0.0510)

1.5 Alpha + real-speech SFT · rank 8

CE 6.46 · genuineness 3.96 · WER 0.0%
Fixed-target emotion MSE 1.197 (normalized 0.0748)

joy disbelief · German · prompt_only

Warte, sag bitte noch nichts; ich habe den ersten Satz jetzt dreimal gelesen und kann es immer noch nicht glauben. Sie bieten mir den Platz an, wirklich diesen Platz mit der kleinen Werkstatt am Meer, von der wir gesprochen haben. Ich habe es geschafft... nach all den Bewerbungen und diesen schrecklichen Gesprächen möchte endlich jemand, dass ich komme. Du bist an jedem zweifelnden Abend bei mir geblieben, und ich möchte, dass du weißt, wie viel mir das bedeutet. Komm her, lass dich kurz umarmen, bevor ich alle anrufe; diesen ersten glücklichen Moment möchte ich nur mit dir teilen.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Disbelief blossoms into exuberant joy and heartfelt gratitude.
SCRIPT:
(very quiet disbelief, rereading the news aloud) [6.4 seconds duration] Warte, sag bitte noch nichts; ich habe den ersten Satz jetzt dreimal gelesen und kann es immer noch nicht glauben.
[0.3 seconds pause]
(hope rising, slightly breathless with excitement) [6.5 seconds duration] Sie bieten mir den Platz an, wirklich diesen Platz mit der kleinen Werkstatt am Meer, von der wir gesprochen haben.
[0.3 seconds pause]
(breathy giggle, 0.4 seconds)
(exuberantly delighted, a spontaneous bright laugh nearby) [6.8 seconds duration] Ich habe es geschafft... nach all den Bewerbungen und diesen schrecklichen Gesprächen möchte endlich jemand, dass ich komme.
[0.3 seconds pause]
(overwhelming gratitude, let the smile soften) [6.8 seconds duration] Du bist an jedem zweifelnden Abend bei mir geblieben, und ich möchte, dass du weißt, wie viel mir das bedeutet.
[0.3 seconds pause]
(warmly elated, sharing rather than announcing) [6.7 seconds duration] Komm her, lass dich kurz umarmen, bevor ich alle anrufe; diesen ersten glücklichen Moment möchte ich nur mit dir teilen.

Fixed seed 20261324 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.41 · genuineness 4.53 · WER 0.0%
Fixed-target emotion MSE 2.523 (normalized 0.1577)

1.5 Alpha · merged, no additional SFT

CE 6.40 · genuineness 4.24 · WER 0.0%
Fixed-target emotion MSE 1.730 (normalized 0.1081)

1.5 Alpha + real-speech SFT · rank 1

CE 6.46 · genuineness 4.30 · WER 0.0%
Fixed-target emotion MSE 1.379 (normalized 0.0862)

1.5 Alpha + real-speech SFT · rank 2

CE 6.45 · genuineness 4.45 · WER 1.0%
Fixed-target emotion MSE 2.311 (normalized 0.1444)

1.5 Alpha + real-speech SFT · rank 4

CE 6.36 · genuineness 4.37 · WER 2.0%
Fixed-target emotion MSE 1.694 (normalized 0.1059)

1.5 Alpha + real-speech SFT · rank 8

CE 6.60 · genuineness 4.55 · WER 0.0%
Fixed-target emotion MSE 2.388 (normalized 0.1492)

joy disbelief · German · reference

Warte, sag bitte noch nichts; ich habe den ersten Satz jetzt dreimal gelesen und kann es immer noch nicht glauben. Sie bieten mir den Platz an, wirklich diesen Platz mit der kleinen Werkstatt am Meer, von der wir gesprochen haben. Ich habe es geschafft... nach all den Bewerbungen und diesen schrecklichen Gesprächen möchte endlich jemand, dass ich komme. Du bist an jedem zweifelnden Abend bei mir geblieben, und ich möchte, dass du weißt, wie viel mir das bedeutet. Komm her, lass dich kurz umarmen, bevor ich alle anrufe; diesen ersten glücklichen Moment möchte ich nur mit dir teilen.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Disbelief blossoms into exuberant joy and heartfelt gratitude.
SCRIPT:
(very quiet disbelief, rereading the news aloud) [6.4 seconds duration] Warte, sag bitte noch nichts; ich habe den ersten Satz jetzt dreimal gelesen und kann es immer noch nicht glauben.
[0.3 seconds pause]
(hope rising, slightly breathless with excitement) [6.5 seconds duration] Sie bieten mir den Platz an, wirklich diesen Platz mit der kleinen Werkstatt am Meer, von der wir gesprochen haben.
[0.3 seconds pause]
(breathy giggle, 0.4 seconds)
(exuberantly delighted, a spontaneous bright laugh nearby) [6.8 seconds duration] Ich habe es geschafft... nach all den Bewerbungen und diesen schrecklichen Gesprächen möchte endlich jemand, dass ich komme.
[0.3 seconds pause]
(overwhelming gratitude, let the smile soften) [6.8 seconds duration] Du bist an jedem zweifelnden Abend bei mir geblieben, und ich möchte, dass du weißt, wie viel mir das bedeutet.
[0.3 seconds pause]
(warmly elated, sharing rather than announcing) [6.7 seconds duration] Komm her, lass dich kurz umarmen, bevor ich alle anrufe; diesen ersten glücklichen Moment möchte ich nur mit dir teilen.

Fixed seed 20261324 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.47 · genuineness 4.53 · WER 2.0%
Fixed-target emotion MSE 1.936 (normalized 0.1210)

1.5 Alpha · merged, no additional SFT

CE 6.40 · genuineness 4.32 · WER 0.0%
Fixed-target emotion MSE 2.190 (normalized 0.1369)

1.5 Alpha + real-speech SFT · rank 1

CE 6.43 · genuineness 4.44 · WER 0.0%
Fixed-target emotion MSE 2.405 (normalized 0.1503)

1.5 Alpha + real-speech SFT · rank 2

CE 6.44 · genuineness 4.46 · WER 2.0%
Fixed-target emotion MSE 2.620 (normalized 0.1638)

1.5 Alpha + real-speech SFT · rank 4

CE 6.47 · genuineness 4.50 · WER 2.0%
Fixed-target emotion MSE 3.004 (normalized 0.1877)

1.5 Alpha + real-speech SFT · rank 8

CE 6.52 · genuineness 4.42 · WER 0.0%
Fixed-target emotion MSE 1.646 (normalized 0.1029)

joy disbelief · Dutch · prompt_only

Wacht, zeg alsjeblieft nog even niets; ik heb die eerste regel nu drie keer gelezen en ik geloof het nog steeds niet. Ze bieden me die plek aan, echt die plek met het kleine atelier aan zee waar we het over hadden. Het is gelukt... na al die sollicitaties en die verschrikkelijke gesprekken wil er eindelijk iemand dat ik kom. Jij bent op al die avonden vol twijfel bij me gebleven, en ik wil dat je weet hoeveel dat betekent. Kom hier, laat me je even vasthouden voordat ik iedereen bel; dit eerste gelukkige moment wil ik met jou delen.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Disbelief blossoms into exuberant joy and heartfelt gratitude.
SCRIPT:
(very quiet disbelief, rereading the news aloud) [6.4 seconds duration] Wacht, zeg alsjeblieft nog even niets; ik heb die eerste regel nu drie keer gelezen en ik geloof het nog steeds niet.
[0.3 seconds pause]
(hope rising, slightly breathless with excitement) [6.5 seconds duration] Ze bieden me die plek aan, echt die plek met het kleine atelier aan zee waar we het over hadden.
[0.3 seconds pause]
(breathy giggle, 0.4 seconds)
(exuberantly delighted, a spontaneous bright laugh nearby) [6.8 seconds duration] Het is gelukt... na al die sollicitaties en die verschrikkelijke gesprekken wil er eindelijk iemand dat ik kom.
[0.3 seconds pause]
(overwhelming gratitude, let the smile soften) [6.8 seconds duration] Jij bent op al die avonden vol twijfel bij me gebleven, en ik wil dat je weet hoeveel dat betekent.
[0.3 seconds pause]
(warmly elated, sharing rather than announcing) [6.7 seconds duration] Kom hier, laat me je even vasthouden voordat ik iedereen bel; dit eerste gelukkige moment wil ik met jou delen.

Fixed seed 20261324 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.31 · genuineness 4.42 · WER 1.0%
Fixed-target emotion MSE 2.982 (normalized 0.1864)

1.5 Alpha · merged, no additional SFT

CE 6.36 · genuineness 4.24 · WER 0.0%
Fixed-target emotion MSE 2.133 (normalized 0.1333)

1.5 Alpha + real-speech SFT · rank 1

CE 6.36 · genuineness 4.24 · WER 0.0%
Fixed-target emotion MSE 3.093 (normalized 0.1933)

1.5 Alpha + real-speech SFT · rank 2

CE 6.36 · genuineness 4.26 · WER 0.0%
Fixed-target emotion MSE 2.480 (normalized 0.1550)

1.5 Alpha + real-speech SFT · rank 4

CE 6.36 · genuineness 4.26 · WER 0.0%
Fixed-target emotion MSE 2.895 (normalized 0.1809)

1.5 Alpha + real-speech SFT · rank 8

CE 6.25 · genuineness 4.14 · WER 0.0%
Fixed-target emotion MSE 3.115 (normalized 0.1947)

joy disbelief · Dutch · reference

Wacht, zeg alsjeblieft nog even niets; ik heb die eerste regel nu drie keer gelezen en ik geloof het nog steeds niet. Ze bieden me die plek aan, echt die plek met het kleine atelier aan zee waar we het over hadden. Het is gelukt... na al die sollicitaties en die verschrikkelijke gesprekken wil er eindelijk iemand dat ik kom. Jij bent op al die avonden vol twijfel bij me gebleven, en ik wil dat je weet hoeveel dat betekent. Kom hier, laat me je even vasthouden voordat ik iedereen bel; dit eerste gelukkige moment wil ik met jou delen.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Disbelief blossoms into exuberant joy and heartfelt gratitude.
SCRIPT:
(very quiet disbelief, rereading the news aloud) [6.4 seconds duration] Wacht, zeg alsjeblieft nog even niets; ik heb die eerste regel nu drie keer gelezen en ik geloof het nog steeds niet.
[0.3 seconds pause]
(hope rising, slightly breathless with excitement) [6.5 seconds duration] Ze bieden me die plek aan, echt die plek met het kleine atelier aan zee waar we het over hadden.
[0.3 seconds pause]
(breathy giggle, 0.4 seconds)
(exuberantly delighted, a spontaneous bright laugh nearby) [6.8 seconds duration] Het is gelukt... na al die sollicitaties en die verschrikkelijke gesprekken wil er eindelijk iemand dat ik kom.
[0.3 seconds pause]
(overwhelming gratitude, let the smile soften) [6.8 seconds duration] Jij bent op al die avonden vol twijfel bij me gebleven, en ik wil dat je weet hoeveel dat betekent.
[0.3 seconds pause]
(warmly elated, sharing rather than announcing) [6.7 seconds duration] Kom hier, laat me je even vasthouden voordat ik iedereen bel; dit eerste gelukkige moment wil ik met jou delen.

Fixed seed 20261324 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.31 · genuineness 4.21 · WER 0.0%
Fixed-target emotion MSE 2.200 (normalized 0.1375)

1.5 Alpha · merged, no additional SFT

CE 6.26 · genuineness 4.22 · WER 1.0%
Fixed-target emotion MSE 3.453 (normalized 0.2158)

1.5 Alpha + real-speech SFT · rank 1

CE 6.25 · genuineness 4.22 · WER 0.0%
Fixed-target emotion MSE 2.051 (normalized 0.1282)

1.5 Alpha + real-speech SFT · rank 2

CE 6.28 · genuineness 4.20 · WER 2.0%
Fixed-target emotion MSE 2.673 (normalized 0.1671)

1.5 Alpha + real-speech SFT · rank 4

CE 6.27 · genuineness 4.16 · WER 0.0%
Fixed-target emotion MSE 2.754 (normalized 0.1721)

1.5 Alpha + real-speech SFT · rank 8

CE 6.28 · genuineness 4.01 · WER 0.0%
Fixed-target emotion MSE 3.404 (normalized 0.2127)

joy disbelief · French · prompt_only

Attends, ne dis rien tout de suite; j'ai relu la première phrase trois fois et je n'arrive toujours pas à y croire. Ils me proposent la place, vraiment cette place avec le petit atelier face à la mer dont nous avions parlé. Ça a marché... après toutes ces candidatures et ces entretiens affreux, quelqu'un veut enfin que je vienne travailler là-bas. Tu es resté près de moi pendant toutes ces soirées de doute, et je veux que tu saches que je n'oublie rien. Viens, laisse-moi te serrer dans mes bras avant d'appeler tout le monde; ce premier moment de bonheur, je le veux avec toi.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Disbelief blossoms into exuberant joy and heartfelt gratitude.
SCRIPT:
(very quiet disbelief, rereading the news aloud) [6.4 seconds duration] Attends, ne dis rien tout de suite; j'ai relu la première phrase trois fois et je n'arrive toujours pas à y croire.
[0.3 seconds pause]
(hope rising, slightly breathless with excitement) [6.5 seconds duration] Ils me proposent la place, vraiment cette place avec le petit atelier face à la mer dont nous avions parlé.
[0.3 seconds pause]
(breathy giggle, 0.4 seconds)
(exuberantly delighted, a spontaneous bright laugh nearby) [6.8 seconds duration] Ça a marché... après toutes ces candidatures et ces entretiens affreux, quelqu'un veut enfin que je vienne travailler là-bas.
[0.3 seconds pause]
(overwhelming gratitude, let the smile soften) [6.8 seconds duration] Tu es resté près de moi pendant toutes ces soirées de doute, et je veux que tu saches que je n'oublie rien.
[0.3 seconds pause]
(warmly elated, sharing rather than announcing) [6.7 seconds duration] Viens, laisse-moi te serrer dans mes bras avant d'appeler tout le monde; ce premier moment de bonheur, je le veux avec toi.

Fixed seed 20261324 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.24 · genuineness 4.37 · WER 20.6%
Fixed-target emotion MSE 1.903 (normalized 0.1189)

1.5 Alpha · merged, no additional SFT

CE 6.30 · genuineness 4.52 · WER 3.7%
Fixed-target emotion MSE 1.367 (normalized 0.0855)

1.5 Alpha + real-speech SFT · rank 1

CE 6.29 · genuineness 4.52 · WER 4.7%
Fixed-target emotion MSE 3.126 (normalized 0.1954)

1.5 Alpha + real-speech SFT · rank 2

CE 6.34 · genuineness 4.63 · WER 2.8%
Fixed-target emotion MSE 2.287 (normalized 0.1429)

1.5 Alpha + real-speech SFT · rank 4

CE 6.34 · genuineness 4.33 · WER 2.8%
Fixed-target emotion MSE 2.380 (normalized 0.1487)

1.5 Alpha + real-speech SFT · rank 8

CE 6.28 · genuineness 4.45 · WER 2.8%
Fixed-target emotion MSE 1.958 (normalized 0.1224)

joy disbelief · French · reference

Attends, ne dis rien tout de suite; j'ai relu la première phrase trois fois et je n'arrive toujours pas à y croire. Ils me proposent la place, vraiment cette place avec le petit atelier face à la mer dont nous avions parlé. Ça a marché... après toutes ces candidatures et ces entretiens affreux, quelqu'un veut enfin que je vienne travailler là-bas. Tu es resté près de moi pendant toutes ces soirées de doute, et je veux que tu saches que je n'oublie rien. Viens, laisse-moi te serrer dans mes bras avant d'appeler tout le monde; ce premier moment de bonheur, je le veux avec toi.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Disbelief blossoms into exuberant joy and heartfelt gratitude.
SCRIPT:
(very quiet disbelief, rereading the news aloud) [6.4 seconds duration] Attends, ne dis rien tout de suite; j'ai relu la première phrase trois fois et je n'arrive toujours pas à y croire.
[0.3 seconds pause]
(hope rising, slightly breathless with excitement) [6.5 seconds duration] Ils me proposent la place, vraiment cette place avec le petit atelier face à la mer dont nous avions parlé.
[0.3 seconds pause]
(breathy giggle, 0.4 seconds)
(exuberantly delighted, a spontaneous bright laugh nearby) [6.8 seconds duration] Ça a marché... après toutes ces candidatures et ces entretiens affreux, quelqu'un veut enfin que je vienne travailler là-bas.
[0.3 seconds pause]
(overwhelming gratitude, let the smile soften) [6.8 seconds duration] Tu es resté près de moi pendant toutes ces soirées de doute, et je veux que tu saches que je n'oublie rien.
[0.3 seconds pause]
(warmly elated, sharing rather than announcing) [6.7 seconds duration] Viens, laisse-moi te serrer dans mes bras avant d'appeler tout le monde; ce premier moment de bonheur, je le veux avec toi.

Fixed seed 20261324 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.27 · genuineness 4.52 · WER 3.7%
Fixed-target emotion MSE 2.680 (normalized 0.1675)

1.5 Alpha · merged, no additional SFT

CE 6.19 · genuineness 4.41 · WER 2.8%
Fixed-target emotion MSE 2.847 (normalized 0.1779)

1.5 Alpha + real-speech SFT · rank 1

CE 6.30 · genuineness 4.57 · WER 3.7%
Fixed-target emotion MSE 1.551 (normalized 0.0970)

1.5 Alpha + real-speech SFT · rank 2

CE 6.16 · genuineness 4.62 · WER 2.8%
Fixed-target emotion MSE 3.536 (normalized 0.2210)

1.5 Alpha + real-speech SFT · rank 4

CE 6.27 · genuineness 4.40 · WER 8.4%
Fixed-target emotion MSE 3.737 (normalized 0.2336)

1.5 Alpha + real-speech SFT · rank 8

CE 6.26 · genuineness 4.48 · WER 5.6%
Fixed-target emotion MSE 2.957 (normalized 0.1848)

joy disbelief · Spanish · prompt_only

Espera, no digas nada todavía; he leído la primera frase tres veces y sigo sin poder creer lo que pone. Me ofrecen la plaza, de verdad, esa plaza con el pequeño taller frente al mar del que tanto habíamos hablado. Lo he conseguido... después de tantas solicitudes y tantas entrevistas horribles, por fin hay alguien que quiere que vaya. Tú te quedaste conmigo todas esas noches llenas de dudas, y necesito que sepas que me acuerdo de cada una. Ven, déjame abrazarte antes de llamar a los demás; quiero que este primer momento de felicidad sea solamente nuestro.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Disbelief blossoms into exuberant joy and heartfelt gratitude.
SCRIPT:
(very quiet disbelief, rereading the news aloud) [6.4 seconds duration] Espera, no digas nada todavía; he leído la primera frase tres veces y sigo sin poder creer lo que pone.
[0.3 seconds pause]
(hope rising, slightly breathless with excitement) [6.5 seconds duration] Me ofrecen la plaza, de verdad, esa plaza con el pequeño taller frente al mar del que tanto habíamos hablado.
[0.3 seconds pause]
(breathy giggle, 0.4 seconds)
(exuberantly delighted, a spontaneous bright laugh nearby) [6.8 seconds duration] Lo he conseguido... después de tantas solicitudes y tantas entrevistas horribles, por fin hay alguien que quiere que vaya.
[0.3 seconds pause]
(overwhelming gratitude, let the smile soften) [6.8 seconds duration] Tú te quedaste conmigo todas esas noches llenas de dudas, y necesito que sepas que me acuerdo de cada una.
[0.3 seconds pause]
(warmly elated, sharing rather than announcing) [6.7 seconds duration] Ven, déjame abrazarte antes de llamar a los demás; quiero que este primer momento de felicidad sea solamente nuestro.

Fixed seed 20261324 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.27 · genuineness 4.42 · WER 0.0%
Fixed-target emotion MSE 2.663 (normalized 0.1664)

1.5 Alpha · merged, no additional SFT

CE 6.13 · genuineness 4.39 · WER 0.0%
Fixed-target emotion MSE 2.079 (normalized 0.1300)

1.5 Alpha + real-speech SFT · rank 1

CE 6.26 · genuineness 4.32 · WER 0.0%
Fixed-target emotion MSE 1.359 (normalized 0.0849)

1.5 Alpha + real-speech SFT · rank 2

CE 6.30 · genuineness 4.25 · WER 0.0%
Fixed-target emotion MSE 2.548 (normalized 0.1593)

1.5 Alpha + real-speech SFT · rank 4

CE 6.22 · genuineness 4.36 · WER 0.0%
Fixed-target emotion MSE 2.021 (normalized 0.1263)

1.5 Alpha + real-speech SFT · rank 8

CE 6.24 · genuineness 4.12 · WER 0.0%
Fixed-target emotion MSE 2.794 (normalized 0.1746)

joy disbelief · Spanish · reference

Espera, no digas nada todavía; he leído la primera frase tres veces y sigo sin poder creer lo que pone. Me ofrecen la plaza, de verdad, esa plaza con el pequeño taller frente al mar del que tanto habíamos hablado. Lo he conseguido... después de tantas solicitudes y tantas entrevistas horribles, por fin hay alguien que quiere que vaya. Tú te quedaste conmigo todas esas noches llenas de dudas, y necesito que sepas que me acuerdo de cada una. Ven, déjame abrazarte antes de llamar a los demás; quiero que este primer momento de felicidad sea solamente nuestro.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Disbelief blossoms into exuberant joy and heartfelt gratitude.
SCRIPT:
(very quiet disbelief, rereading the news aloud) [6.4 seconds duration] Espera, no digas nada todavía; he leído la primera frase tres veces y sigo sin poder creer lo que pone.
[0.3 seconds pause]
(hope rising, slightly breathless with excitement) [6.5 seconds duration] Me ofrecen la plaza, de verdad, esa plaza con el pequeño taller frente al mar del que tanto habíamos hablado.
[0.3 seconds pause]
(breathy giggle, 0.4 seconds)
(exuberantly delighted, a spontaneous bright laugh nearby) [6.8 seconds duration] Lo he conseguido... después de tantas solicitudes y tantas entrevistas horribles, por fin hay alguien que quiere que vaya.
[0.3 seconds pause]
(overwhelming gratitude, let the smile soften) [6.8 seconds duration] Tú te quedaste conmigo todas esas noches llenas de dudas, y necesito que sepas que me acuerdo de cada una.
[0.3 seconds pause]
(warmly elated, sharing rather than announcing) [6.7 seconds duration] Ven, déjame abrazarte antes de llamar a los demás; quiero que este primer momento de felicidad sea solamente nuestro.

Fixed seed 20261324 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.26 · genuineness 4.26 · WER 0.0%
Fixed-target emotion MSE 0.688 (normalized 0.0430)

1.5 Alpha · merged, no additional SFT

CE 6.17 · genuineness 4.28 · WER 0.0%
Fixed-target emotion MSE 2.042 (normalized 0.1276)

1.5 Alpha + real-speech SFT · rank 1

CE 6.25 · genuineness 4.35 · WER 0.0%
Fixed-target emotion MSE 1.491 (normalized 0.0932)

1.5 Alpha + real-speech SFT · rank 2

CE 6.16 · genuineness 4.06 · WER 0.0%
Fixed-target emotion MSE 2.051 (normalized 0.1282)

1.5 Alpha + real-speech SFT · rank 4

CE 6.16 · genuineness 4.05 · WER 0.0%
Fixed-target emotion MSE 2.077 (normalized 0.1298)

1.5 Alpha + real-speech SFT · rank 8

CE 6.15 · genuineness 4.29 · WER 1.0%
Fixed-target emotion MSE 2.123 (normalized 0.1327)

amusement embarrassment · English · prompt_only

So I walked into the meeting feeling very professional, carrying my notes and the expensive coffee I had bought to look organized. Everyone went quiet when I connected the laptop, and I honestly thought they were impressed by how prepared I was. Then I turned around... my entire presentation was the photograph of your cat wearing that enormous orange birthday hat. I tried to explain, but the director asked whether our new strategy involved more naps, and that was the end of me. Anyway, your cat is apparently our unofficial consultant now, so please tell him I expect a very reasonable invoice.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Conversational storytelling turns into embarrassed laughter and affectionate amusement.
SCRIPT:
(casual storytelling, confidently amused) [6.4 seconds duration] So I walked into the meeting feeling very professional, carrying my notes and the expensive coffee I had bought to look organized.
[0.3 seconds pause]
(playful suspense, trying to remain serious) [6.5 seconds duration] Everyone went quiet when I connected the laptop, and I honestly thought they were impressed by how prepared I was.
[0.3 seconds pause]
(chuckle, 0.4 seconds)
(embarrassed realization, a tiny hesitation) [6.8 seconds duration] Then I turned around... my entire presentation was the photograph of your cat wearing that enormous orange birthday hat.
[0.3 seconds pause]
(warm unrestrained amusement, laughing through the memory) [6.8 seconds duration] I tried to explain, but the director asked whether our new strategy involved more naps, and that was the end of me.
[0.3 seconds pause]
(affectionate and self-deprecating, winding down naturally) [6.7 seconds duration] Anyway, your cat is apparently our unofficial consultant now, so please tell him I expect a very reasonable invoice.

Fixed seed 20261425 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.45 · genuineness 4.33 · WER 2.9%
Fixed-target emotion MSE 0.286 (normalized 0.0179)

1.5 Alpha · merged, no additional SFT

CE 6.52 · genuineness 4.43 · WER 2.0%
Fixed-target emotion MSE 1.043 (normalized 0.0652)

1.5 Alpha + real-speech SFT · rank 1

CE 6.52 · genuineness 4.20 · WER 0.0%
Fixed-target emotion MSE 0.621 (normalized 0.0388)

1.5 Alpha + real-speech SFT · rank 2

CE 6.51 · genuineness 4.49 · WER 2.9%
Fixed-target emotion MSE 0.893 (normalized 0.0558)

1.5 Alpha + real-speech SFT · rank 4

CE 6.57 · genuineness 4.53 · WER 0.0%
Fixed-target emotion MSE 0.835 (normalized 0.0522)

1.5 Alpha + real-speech SFT · rank 8

CE 6.54 · genuineness 4.39 · WER 2.0%
Fixed-target emotion MSE 0.891 (normalized 0.0557)

amusement embarrassment · English · reference

So I walked into the meeting feeling very professional, carrying my notes and the expensive coffee I had bought to look organized. Everyone went quiet when I connected the laptop, and I honestly thought they were impressed by how prepared I was. Then I turned around... my entire presentation was the photograph of your cat wearing that enormous orange birthday hat. I tried to explain, but the director asked whether our new strategy involved more naps, and that was the end of me. Anyway, your cat is apparently our unofficial consultant now, so please tell him I expect a very reasonable invoice.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Conversational storytelling turns into embarrassed laughter and affectionate amusement.
SCRIPT:
(casual storytelling, confidently amused) [6.4 seconds duration] So I walked into the meeting feeling very professional, carrying my notes and the expensive coffee I had bought to look organized.
[0.3 seconds pause]
(playful suspense, trying to remain serious) [6.5 seconds duration] Everyone went quiet when I connected the laptop, and I honestly thought they were impressed by how prepared I was.
[0.3 seconds pause]
(chuckle, 0.4 seconds)
(embarrassed realization, a tiny hesitation) [6.8 seconds duration] Then I turned around... my entire presentation was the photograph of your cat wearing that enormous orange birthday hat.
[0.3 seconds pause]
(warm unrestrained amusement, laughing through the memory) [6.8 seconds duration] I tried to explain, but the director asked whether our new strategy involved more naps, and that was the end of me.
[0.3 seconds pause]
(affectionate and self-deprecating, winding down naturally) [6.7 seconds duration] Anyway, your cat is apparently our unofficial consultant now, so please tell him I expect a very reasonable invoice.

Fixed seed 20261425 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.48 · genuineness 4.08 · WER 2.0%
Fixed-target emotion MSE 0.877 (normalized 0.0548)

1.5 Alpha · merged, no additional SFT

CE 6.50 · genuineness 4.24 · WER 0.0%
Fixed-target emotion MSE 0.975 (normalized 0.0609)

1.5 Alpha + real-speech SFT · rank 1

CE 6.46 · genuineness 4.17 · WER 0.0%
Fixed-target emotion MSE 1.168 (normalized 0.0730)

1.5 Alpha + real-speech SFT · rank 2

CE 6.48 · genuineness 4.17 · WER 0.0%
Fixed-target emotion MSE 1.043 (normalized 0.0652)

1.5 Alpha + real-speech SFT · rank 4

CE 6.43 · genuineness 4.01 · WER 0.0%
Fixed-target emotion MSE 1.345 (normalized 0.0841)

1.5 Alpha + real-speech SFT · rank 8

CE 6.45 · genuineness 4.01 · WER 0.0%
Fixed-target emotion MSE 1.302 (normalized 0.0814)

amusement embarrassment · German · prompt_only

Also, ich kam ziemlich selbstbewusst in die Besprechung, mit meinen Notizen und diesem teuren Kaffee, damit alles schön organisiert aussieht. Als ich den Laptop anschloss, wurden plötzlich alle still, und ich dachte tatsächlich, sie wären von meiner Vorbereitung beeindruckt. Dann habe ich mich umgedreht... auf der ganzen Leinwand war dein Kater mit diesem riesigen orangefarbenen Geburtstagshut zu sehen. Ich wollte es erklären, aber der Chef fragte, ob unsere neue Strategie mehr Mittagsschlaf vorsieht, und da war es vorbei. Jedenfalls ist dein Kater jetzt unser inoffizieller Berater; sag ihm bitte, dass ich eine wirklich faire Rechnung erwarte.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Conversational storytelling turns into embarrassed laughter and affectionate amusement.
SCRIPT:
(casual storytelling, confidently amused) [6.4 seconds duration] Also, ich kam ziemlich selbstbewusst in die Besprechung, mit meinen Notizen und diesem teuren Kaffee, damit alles schön organisiert aussieht.
[0.3 seconds pause]
(playful suspense, trying to remain serious) [6.5 seconds duration] Als ich den Laptop anschloss, wurden plötzlich alle still, und ich dachte tatsächlich, sie wären von meiner Vorbereitung beeindruckt.
[0.3 seconds pause]
(chuckle, 0.4 seconds)
(embarrassed realization, a tiny hesitation) [6.8 seconds duration] Dann habe ich mich umgedreht... auf der ganzen Leinwand war dein Kater mit diesem riesigen orangefarbenen Geburtstagshut zu sehen.
[0.3 seconds pause]
(warm unrestrained amusement, laughing through the memory) [6.8 seconds duration] Ich wollte es erklären, aber der Chef fragte, ob unsere neue Strategie mehr Mittagsschlaf vorsieht, und da war es vorbei.
[0.3 seconds pause]
(affectionate and self-deprecating, winding down naturally) [6.7 seconds duration] Jedenfalls ist dein Kater jetzt unser inoffizieller Berater; sag ihm bitte, dass ich eine wirklich faire Rechnung erwarte.

Fixed seed 20261425 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.42 · genuineness 4.32 · WER 3.1%
Fixed-target emotion MSE 1.775 (normalized 0.1109)

1.5 Alpha · merged, no additional SFT

CE 6.38 · genuineness 4.53 · WER 2.1%
Fixed-target emotion MSE 1.658 (normalized 0.1036)

1.5 Alpha + real-speech SFT · rank 1

CE 6.32 · genuineness 4.49 · WER 0.0%
Fixed-target emotion MSE 1.172 (normalized 0.0733)

1.5 Alpha + real-speech SFT · rank 2

CE 6.47 · genuineness 4.46 · WER 1.0%
Fixed-target emotion MSE 1.017 (normalized 0.0636)

1.5 Alpha + real-speech SFT · rank 4

CE 6.40 · genuineness 4.36 · WER 2.1%
Fixed-target emotion MSE 1.163 (normalized 0.0727)

1.5 Alpha + real-speech SFT · rank 8

CE 6.34 · genuineness 4.46 · WER 1.0%
Fixed-target emotion MSE 1.701 (normalized 0.1063)

amusement embarrassment · German · reference

Also, ich kam ziemlich selbstbewusst in die Besprechung, mit meinen Notizen und diesem teuren Kaffee, damit alles schön organisiert aussieht. Als ich den Laptop anschloss, wurden plötzlich alle still, und ich dachte tatsächlich, sie wären von meiner Vorbereitung beeindruckt. Dann habe ich mich umgedreht... auf der ganzen Leinwand war dein Kater mit diesem riesigen orangefarbenen Geburtstagshut zu sehen. Ich wollte es erklären, aber der Chef fragte, ob unsere neue Strategie mehr Mittagsschlaf vorsieht, und da war es vorbei. Jedenfalls ist dein Kater jetzt unser inoffizieller Berater; sag ihm bitte, dass ich eine wirklich faire Rechnung erwarte.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Conversational storytelling turns into embarrassed laughter and affectionate amusement.
SCRIPT:
(casual storytelling, confidently amused) [6.4 seconds duration] Also, ich kam ziemlich selbstbewusst in die Besprechung, mit meinen Notizen und diesem teuren Kaffee, damit alles schön organisiert aussieht.
[0.3 seconds pause]
(playful suspense, trying to remain serious) [6.5 seconds duration] Als ich den Laptop anschloss, wurden plötzlich alle still, und ich dachte tatsächlich, sie wären von meiner Vorbereitung beeindruckt.
[0.3 seconds pause]
(chuckle, 0.4 seconds)
(embarrassed realization, a tiny hesitation) [6.8 seconds duration] Dann habe ich mich umgedreht... auf der ganzen Leinwand war dein Kater mit diesem riesigen orangefarbenen Geburtstagshut zu sehen.
[0.3 seconds pause]
(warm unrestrained amusement, laughing through the memory) [6.8 seconds duration] Ich wollte es erklären, aber der Chef fragte, ob unsere neue Strategie mehr Mittagsschlaf vorsieht, und da war es vorbei.
[0.3 seconds pause]
(affectionate and self-deprecating, winding down naturally) [6.7 seconds duration] Jedenfalls ist dein Kater jetzt unser inoffizieller Berater; sag ihm bitte, dass ich eine wirklich faire Rechnung erwarte.

Fixed seed 20261425 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.44 · genuineness 4.28 · WER 2.1%
Fixed-target emotion MSE 1.178 (normalized 0.0736)

1.5 Alpha · merged, no additional SFT

CE 6.47 · genuineness 4.42 · WER 4.2%
Fixed-target emotion MSE 1.327 (normalized 0.0829)

1.5 Alpha + real-speech SFT · rank 1

CE 6.44 · genuineness 4.37 · WER 0.0%
Fixed-target emotion MSE 0.993 (normalized 0.0621)

1.5 Alpha + real-speech SFT · rank 2

CE 6.47 · genuineness 4.36 · WER 5.2%
Fixed-target emotion MSE 1.524 (normalized 0.0952)

1.5 Alpha + real-speech SFT · rank 4

CE 6.45 · genuineness 4.56 · WER 2.1%
Fixed-target emotion MSE 1.260 (normalized 0.0788)

1.5 Alpha + real-speech SFT · rank 8

CE 6.42 · genuineness 4.27 · WER 1.0%
Fixed-target emotion MSE 1.584 (normalized 0.0990)

amusement embarrassment · Dutch · prompt_only

Dus ik liep heel zelfverzekerd die vergadering binnen, met mijn notities en dure koffie, zodat ik lekker georganiseerd zou lijken. Toen ik mijn laptop aansloot, werd iedereen ineens stil, en ik dacht werkelijk dat ze onder de indruk waren. Toen draaide ik me om... jouw kat stond levensgroot op het scherm, met die belachelijk grote oranje verjaardagshoed op. Ik wilde het uitleggen, maar de directeur vroeg of onze nieuwe strategie meer middagdutjes inhield, en toen hield ik het niet meer. Hoe dan ook, jouw kat is nu onze onofficiële adviseur, dus zeg hem alsjeblieft dat ik een redelijke rekening verwacht.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Conversational storytelling turns into embarrassed laughter and affectionate amusement.
SCRIPT:
(casual storytelling, confidently amused) [6.4 seconds duration] Dus ik liep heel zelfverzekerd die vergadering binnen, met mijn notities en dure koffie, zodat ik lekker georganiseerd zou lijken.
[0.3 seconds pause]
(playful suspense, trying to remain serious) [6.5 seconds duration] Toen ik mijn laptop aansloot, werd iedereen ineens stil, en ik dacht werkelijk dat ze onder de indruk waren.
[0.3 seconds pause]
(chuckle, 0.4 seconds)
(embarrassed realization, a tiny hesitation) [6.8 seconds duration] Toen draaide ik me om... jouw kat stond levensgroot op het scherm, met die belachelijk grote oranje verjaardagshoed op.
[0.3 seconds pause]
(warm unrestrained amusement, laughing through the memory) [6.8 seconds duration] Ik wilde het uitleggen, maar de directeur vroeg of onze nieuwe strategie meer middagdutjes inhield, en toen hield ik het niet meer.
[0.3 seconds pause]
(affectionate and self-deprecating, winding down naturally) [6.7 seconds duration] Hoe dan ook, jouw kat is nu onze onofficiële adviseur, dus zeg hem alsjeblieft dat ik een redelijke rekening verwacht.

Fixed seed 20261425 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.18 · genuineness 4.31 · WER 3.0%
Fixed-target emotion MSE 1.235 (normalized 0.0772)

1.5 Alpha · merged, no additional SFT

CE 6.33 · genuineness 4.20 · WER 4.0%
Fixed-target emotion MSE 1.635 (normalized 0.1022)

1.5 Alpha + real-speech SFT · rank 1

CE 6.33 · genuineness 4.24 · WER 1.0%
Fixed-target emotion MSE 1.257 (normalized 0.0786)

1.5 Alpha + real-speech SFT · rank 2

CE 6.25 · genuineness 4.14 · WER 0.0%
Fixed-target emotion MSE 1.490 (normalized 0.0932)

1.5 Alpha + real-speech SFT · rank 4

CE 6.32 · genuineness 4.17 · WER 1.0%
Fixed-target emotion MSE 1.537 (normalized 0.0961)

1.5 Alpha + real-speech SFT · rank 8

CE 6.26 · genuineness 4.08 · WER 5.0%
Fixed-target emotion MSE 1.519 (normalized 0.0949)

amusement embarrassment · Dutch · reference

Dus ik liep heel zelfverzekerd die vergadering binnen, met mijn notities en dure koffie, zodat ik lekker georganiseerd zou lijken. Toen ik mijn laptop aansloot, werd iedereen ineens stil, en ik dacht werkelijk dat ze onder de indruk waren. Toen draaide ik me om... jouw kat stond levensgroot op het scherm, met die belachelijk grote oranje verjaardagshoed op. Ik wilde het uitleggen, maar de directeur vroeg of onze nieuwe strategie meer middagdutjes inhield, en toen hield ik het niet meer. Hoe dan ook, jouw kat is nu onze onofficiële adviseur, dus zeg hem alsjeblieft dat ik een redelijke rekening verwacht.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Conversational storytelling turns into embarrassed laughter and affectionate amusement.
SCRIPT:
(casual storytelling, confidently amused) [6.4 seconds duration] Dus ik liep heel zelfverzekerd die vergadering binnen, met mijn notities en dure koffie, zodat ik lekker georganiseerd zou lijken.
[0.3 seconds pause]
(playful suspense, trying to remain serious) [6.5 seconds duration] Toen ik mijn laptop aansloot, werd iedereen ineens stil, en ik dacht werkelijk dat ze onder de indruk waren.
[0.3 seconds pause]
(chuckle, 0.4 seconds)
(embarrassed realization, a tiny hesitation) [6.8 seconds duration] Toen draaide ik me om... jouw kat stond levensgroot op het scherm, met die belachelijk grote oranje verjaardagshoed op.
[0.3 seconds pause]
(warm unrestrained amusement, laughing through the memory) [6.8 seconds duration] Ik wilde het uitleggen, maar de directeur vroeg of onze nieuwe strategie meer middagdutjes inhield, en toen hield ik het niet meer.
[0.3 seconds pause]
(affectionate and self-deprecating, winding down naturally) [6.7 seconds duration] Hoe dan ook, jouw kat is nu onze onofficiële adviseur, dus zeg hem alsjeblieft dat ik een redelijke rekening verwacht.

Fixed seed 20261425 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.20 · genuineness 4.09 · WER 5.0%
Fixed-target emotion MSE 1.809 (normalized 0.1130)

1.5 Alpha · merged, no additional SFT

CE 6.29 · genuineness 4.12 · WER 1.0%
Fixed-target emotion MSE 2.039 (normalized 0.1275)

1.5 Alpha + real-speech SFT · rank 1

CE 6.22 · genuineness 4.19 · WER 2.0%
Fixed-target emotion MSE 2.123 (normalized 0.1327)

1.5 Alpha + real-speech SFT · rank 2

CE 6.23 · genuineness 4.10 · WER 3.0%
Fixed-target emotion MSE 1.863 (normalized 0.1165)

1.5 Alpha + real-speech SFT · rank 4

CE 6.21 · genuineness 3.93 · WER 0.0%
Fixed-target emotion MSE 1.593 (normalized 0.0996)

1.5 Alpha + real-speech SFT · rank 8

CE 6.26 · genuineness 4.09 · WER 3.0%
Fixed-target emotion MSE 1.535 (normalized 0.0960)

amusement embarrassment · French · prompt_only

Alors, je suis arrivé à la réunion avec mes notes et un café beaucoup trop cher pour avoir l'air parfaitement organisé. Tout le monde s'est tu quand j'ai branché mon ordinateur, et j'ai vraiment cru que ma préparation les impressionnait. Puis je me suis retourné... sur tout l'écran, il y avait ton chat avec cet immense chapeau d'anniversaire orange. J'ai voulu expliquer, mais la directrice a demandé si notre nouvelle stratégie prévoyait davantage de siestes, et là, j'ai craqué. Bref, ton chat est devenu notre conseiller officieux; dis-lui quand même que j'espère recevoir une facture tout à fait raisonnable.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Conversational storytelling turns into embarrassed laughter and affectionate amusement.
SCRIPT:
(casual storytelling, confidently amused) [6.4 seconds duration] Alors, je suis arrivé à la réunion avec mes notes et un café beaucoup trop cher pour avoir l'air parfaitement organisé.
[0.3 seconds pause]
(playful suspense, trying to remain serious) [6.5 seconds duration] Tout le monde s'est tu quand j'ai branché mon ordinateur, et j'ai vraiment cru que ma préparation les impressionnait.
[0.3 seconds pause]
(chuckle, 0.4 seconds)
(embarrassed realization, a tiny hesitation) [6.8 seconds duration] Puis je me suis retourné... sur tout l'écran, il y avait ton chat avec cet immense chapeau d'anniversaire orange.
[0.3 seconds pause]
(warm unrestrained amusement, laughing through the memory) [6.8 seconds duration] J'ai voulu expliquer, mais la directrice a demandé si notre nouvelle stratégie prévoyait davantage de siestes, et là, j'ai craqué.
[0.3 seconds pause]
(affectionate and self-deprecating, winding down naturally) [6.7 seconds duration] Bref, ton chat est devenu notre conseiller officieux; dis-lui quand même que j'espère recevoir une facture tout à fait raisonnable.

Fixed seed 20261425 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.25 · genuineness 4.63 · WER 6.0%
Fixed-target emotion MSE 1.473 (normalized 0.0921)

1.5 Alpha · merged, no additional SFT

CE 6.15 · genuineness 4.56 · WER 17.0%
Fixed-target emotion MSE 1.236 (normalized 0.0773)

1.5 Alpha + real-speech SFT · rank 1

CE 6.15 · genuineness 4.31 · WER 82.0%
Fixed-target emotion MSE 1.860 (normalized 0.1163)

1.5 Alpha + real-speech SFT · rank 2

CE 6.16 · genuineness 4.39 · WER 4.0%
Fixed-target emotion MSE 2.462 (normalized 0.1539)

1.5 Alpha + real-speech SFT · rank 4

CE 6.08 · genuineness 4.17 · WER 4.0%
Fixed-target emotion MSE 1.688 (normalized 0.1055)

1.5 Alpha + real-speech SFT · rank 8

CE 6.18 · genuineness 4.30 · WER 5.0%
Fixed-target emotion MSE 2.265 (normalized 0.1416)

amusement embarrassment · French · reference

Alors, je suis arrivé à la réunion avec mes notes et un café beaucoup trop cher pour avoir l'air parfaitement organisé. Tout le monde s'est tu quand j'ai branché mon ordinateur, et j'ai vraiment cru que ma préparation les impressionnait. Puis je me suis retourné... sur tout l'écran, il y avait ton chat avec cet immense chapeau d'anniversaire orange. J'ai voulu expliquer, mais la directrice a demandé si notre nouvelle stratégie prévoyait davantage de siestes, et là, j'ai craqué. Bref, ton chat est devenu notre conseiller officieux; dis-lui quand même que j'espère recevoir une facture tout à fait raisonnable.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Conversational storytelling turns into embarrassed laughter and affectionate amusement.
SCRIPT:
(casual storytelling, confidently amused) [6.4 seconds duration] Alors, je suis arrivé à la réunion avec mes notes et un café beaucoup trop cher pour avoir l'air parfaitement organisé.
[0.3 seconds pause]
(playful suspense, trying to remain serious) [6.5 seconds duration] Tout le monde s'est tu quand j'ai branché mon ordinateur, et j'ai vraiment cru que ma préparation les impressionnait.
[0.3 seconds pause]
(chuckle, 0.4 seconds)
(embarrassed realization, a tiny hesitation) [6.8 seconds duration] Puis je me suis retourné... sur tout l'écran, il y avait ton chat avec cet immense chapeau d'anniversaire orange.
[0.3 seconds pause]
(warm unrestrained amusement, laughing through the memory) [6.8 seconds duration] J'ai voulu expliquer, mais la directrice a demandé si notre nouvelle stratégie prévoyait davantage de siestes, et là, j'ai craqué.
[0.3 seconds pause]
(affectionate and self-deprecating, winding down naturally) [6.7 seconds duration] Bref, ton chat est devenu notre conseiller officieux; dis-lui quand même que j'espère recevoir une facture tout à fait raisonnable.

Fixed seed 20261425 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.26 · genuineness 4.37 · WER 2.0%
Fixed-target emotion MSE 2.452 (normalized 0.1532)

1.5 Alpha · merged, no additional SFT

CE 6.05 · genuineness 4.58 · WER 4.0%
Fixed-target emotion MSE 2.714 (normalized 0.1696)

1.5 Alpha + real-speech SFT · rank 1

CE 6.17 · genuineness 4.40 · WER 3.0%
Fixed-target emotion MSE 2.183 (normalized 0.1364)

1.5 Alpha + real-speech SFT · rank 2

CE 6.18 · genuineness 4.44 · WER 3.0%
Fixed-target emotion MSE 2.405 (normalized 0.1503)

1.5 Alpha + real-speech SFT · rank 4

CE 6.01 · genuineness 4.57 · WER 2.0%
Fixed-target emotion MSE 3.118 (normalized 0.1949)

1.5 Alpha + real-speech SFT · rank 8

CE 6.22 · genuineness 4.48 · WER 4.0%
Fixed-target emotion MSE 2.013 (normalized 0.1258)

amusement embarrassment · Spanish · prompt_only

Pues entré en la reunión sintiéndome muy profesional, con mis notas y un café carísimo para parecer una persona organizada. Todos se quedaron callados cuando conecté el portátil, y yo pensé de verdad que estaban impresionados con mi preparación. Entonces me di la vuelta... toda la pantalla mostraba a tu gato llevando aquel enorme sombrero naranja de cumpleaños. Intenté explicarlo, pero la directora preguntó si nuestra nueva estrategia incluía más siestas, y ahí ya no pude aguantarme. En fin, tu gato es ahora nuestro asesor extraoficial; dile, por favor, que espero una factura bastante razonable por sus servicios.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Conversational storytelling turns into embarrassed laughter and affectionate amusement.
SCRIPT:
(casual storytelling, confidently amused) [6.4 seconds duration] Pues entré en la reunión sintiéndome muy profesional, con mis notas y un café carísimo para parecer una persona organizada.
[0.3 seconds pause]
(playful suspense, trying to remain serious) [6.5 seconds duration] Todos se quedaron callados cuando conecté el portátil, y yo pensé de verdad que estaban impresionados con mi preparación.
[0.3 seconds pause]
(chuckle, 0.4 seconds)
(embarrassed realization, a tiny hesitation) [6.8 seconds duration] Entonces me di la vuelta... toda la pantalla mostraba a tu gato llevando aquel enorme sombrero naranja de cumpleaños.
[0.3 seconds pause]
(warm unrestrained amusement, laughing through the memory) [6.8 seconds duration] Intenté explicarlo, pero la directora preguntó si nuestra nueva estrategia incluía más siestas, y ahí ya no pude aguantarme.
[0.3 seconds pause]
(affectionate and self-deprecating, winding down naturally) [6.7 seconds duration] En fin, tu gato es ahora nuestro asesor extraoficial; dile, por favor, que espero una factura bastante razonable por sus servicios.

Fixed seed 20261425 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.16 · genuineness 4.15 · WER 0.0%
Fixed-target emotion MSE 0.972 (normalized 0.0607)

1.5 Alpha · merged, no additional SFT

CE 6.21 · genuineness 4.31 · WER 2.0%
Fixed-target emotion MSE 2.331 (normalized 0.1457)

1.5 Alpha + real-speech SFT · rank 1

CE 6.28 · genuineness 4.04 · WER 2.0%
Fixed-target emotion MSE 1.664 (normalized 0.1040)

1.5 Alpha + real-speech SFT · rank 2

CE 6.23 · genuineness 4.26 · WER 1.0%
Fixed-target emotion MSE 0.963 (normalized 0.0602)

1.5 Alpha + real-speech SFT · rank 4

CE 6.30 · genuineness 4.24 · WER 1.0%
Fixed-target emotion MSE 1.174 (normalized 0.0734)

1.5 Alpha + real-speech SFT · rank 8

CE 6.25 · genuineness 4.28 · WER 0.0%
Fixed-target emotion MSE 1.550 (normalized 0.0969)

amusement embarrassment · Spanish · reference

Pues entré en la reunión sintiéndome muy profesional, con mis notas y un café carísimo para parecer una persona organizada. Todos se quedaron callados cuando conecté el portátil, y yo pensé de verdad que estaban impresionados con mi preparación. Entonces me di la vuelta... toda la pantalla mostraba a tu gato llevando aquel enorme sombrero naranja de cumpleaños. Intenté explicarlo, pero la directora preguntó si nuestra nueva estrategia incluía más siestas, y ahí ya no pude aguantarme. En fin, tu gato es ahora nuestro asesor extraoficial; dile, por favor, que espero una factura bastante razonable por sus servicios.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Conversational storytelling turns into embarrassed laughter and affectionate amusement.
SCRIPT:
(casual storytelling, confidently amused) [6.4 seconds duration] Pues entré en la reunión sintiéndome muy profesional, con mis notas y un café carísimo para parecer una persona organizada.
[0.3 seconds pause]
(playful suspense, trying to remain serious) [6.5 seconds duration] Todos se quedaron callados cuando conecté el portátil, y yo pensé de verdad que estaban impresionados con mi preparación.
[0.3 seconds pause]
(chuckle, 0.4 seconds)
(embarrassed realization, a tiny hesitation) [6.8 seconds duration] Entonces me di la vuelta... toda la pantalla mostraba a tu gato llevando aquel enorme sombrero naranja de cumpleaños.
[0.3 seconds pause]
(warm unrestrained amusement, laughing through the memory) [6.8 seconds duration] Intenté explicarlo, pero la directora preguntó si nuestra nueva estrategia incluía más siestas, y ahí ya no pude aguantarme.
[0.3 seconds pause]
(affectionate and self-deprecating, winding down naturally) [6.7 seconds duration] En fin, tu gato es ahora nuestro asesor extraoficial; dile, por favor, que espero una factura bastante razonable por sus servicios.

Fixed seed 20261425 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.23 · genuineness 4.21 · WER 3.1%
Fixed-target emotion MSE 0.414 (normalized 0.0259)

1.5 Alpha · merged, no additional SFT

CE 6.09 · genuineness 3.92 · WER 1.0%
Fixed-target emotion MSE 1.580 (normalized 0.0988)

1.5 Alpha + real-speech SFT · rank 1

CE 6.19 · genuineness 3.92 · WER 0.0%
Fixed-target emotion MSE 0.814 (normalized 0.0509)

1.5 Alpha + real-speech SFT · rank 2

CE 6.12 · genuineness 4.17 · WER 0.0%
Fixed-target emotion MSE 1.425 (normalized 0.0891)

1.5 Alpha + real-speech SFT · rank 4

CE 6.25 · genuineness 3.87 · WER 0.0%
Fixed-target emotion MSE 1.751 (normalized 0.1094)

1.5 Alpha + real-speech SFT · rank 8

CE 6.13 · genuineness 4.04 · WER 0.0%
Fixed-target emotion MSE 1.292 (normalized 0.0807)

betrayal resolve · English · prompt_only

I saw the announcement this morning, and at first I thought there must be another page with the rest of our names. But there was only yours, underneath the idea we stayed here building together while everyone else had already gone home. You even used the little phrase I wrote on the window... that is almost funny, in a really miserable sort of way. I am not asking you to apologize beautifully; I am asking you to correct the announcement and give the team its credit. After that, we can talk about working together again, but not before you understand why I cannot simply smile this away.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Polite restraint fractures into hurt and bitterness, then becomes self-respecting resolve.
SCRIPT:
(carefully polite, hurt held beneath the surface) [6.4 seconds duration] I saw the announcement this morning, and at first I thought there must be another page with the rest of our names.
[0.3 seconds pause]
(disbelief, a little colder and slower) [6.5 seconds duration] But there was only yours, underneath the idea we stayed here building together while everyone else had already gone home.
[0.3 seconds pause]
(low bitter chuckle, 0.4 seconds)
(bitter and wounded, not villainous) [6.8 seconds duration] You even used the little phrase I wrote on the window... that is almost funny, in a really miserable sort of way.
[0.3 seconds pause]
(firm, controlled anger, clearly articulated) [6.8 seconds duration] I am not asking you to apologize beautifully; I am asking you to correct the announcement and give the team its credit.
[0.3 seconds pause]
(quiet dignity, a boundary stated without shouting) [6.7 seconds duration] After that, we can talk about working together again, but not before you understand why I cannot simply smile this away.

Fixed seed 20261526 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.58 · genuineness 4.47 · WER 0.9%
Fixed-target emotion MSE 2.206 (normalized 0.1379)

1.5 Alpha · merged, no additional SFT

CE 6.53 · genuineness 4.02 · WER 0.0%
Fixed-target emotion MSE 2.756 (normalized 0.1723)

1.5 Alpha + real-speech SFT · rank 1

CE 6.61 · genuineness 4.30 · WER 0.0%
Fixed-target emotion MSE 1.909 (normalized 0.1193)

1.5 Alpha + real-speech SFT · rank 2

CE 6.55 · genuineness 4.21 · WER 0.0%
Fixed-target emotion MSE 2.635 (normalized 0.1647)

1.5 Alpha + real-speech SFT · rank 4

CE 6.59 · genuineness 4.19 · WER 0.0%
Fixed-target emotion MSE 2.541 (normalized 0.1588)

1.5 Alpha + real-speech SFT · rank 8

CE 6.61 · genuineness 4.33 · WER 0.0%
Fixed-target emotion MSE 2.309 (normalized 0.1443)

betrayal resolve · English · reference

I saw the announcement this morning, and at first I thought there must be another page with the rest of our names. But there was only yours, underneath the idea we stayed here building together while everyone else had already gone home. You even used the little phrase I wrote on the window... that is almost funny, in a really miserable sort of way. I am not asking you to apologize beautifully; I am asking you to correct the announcement and give the team its credit. After that, we can talk about working together again, but not before you understand why I cannot simply smile this away.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Polite restraint fractures into hurt and bitterness, then becomes self-respecting resolve.
SCRIPT:
(carefully polite, hurt held beneath the surface) [6.4 seconds duration] I saw the announcement this morning, and at first I thought there must be another page with the rest of our names.
[0.3 seconds pause]
(disbelief, a little colder and slower) [6.5 seconds duration] But there was only yours, underneath the idea we stayed here building together while everyone else had already gone home.
[0.3 seconds pause]
(low bitter chuckle, 0.4 seconds)
(bitter and wounded, not villainous) [6.8 seconds duration] You even used the little phrase I wrote on the window... that is almost funny, in a really miserable sort of way.
[0.3 seconds pause]
(firm, controlled anger, clearly articulated) [6.8 seconds duration] I am not asking you to apologize beautifully; I am asking you to correct the announcement and give the team its credit.
[0.3 seconds pause]
(quiet dignity, a boundary stated without shouting) [6.7 seconds duration] After that, we can talk about working together again, but not before you understand why I cannot simply smile this away.

Fixed seed 20261526 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.50 · genuineness 3.96 · WER 0.9%
Fixed-target emotion MSE 2.751 (normalized 0.1720)

1.5 Alpha · merged, no additional SFT

CE 6.44 · genuineness 3.88 · WER 0.9%
Fixed-target emotion MSE 1.816 (normalized 0.1135)

1.5 Alpha + real-speech SFT · rank 1

CE 6.49 · genuineness 3.92 · WER 0.9%
Fixed-target emotion MSE 2.443 (normalized 0.1527)

1.5 Alpha + real-speech SFT · rank 2

CE 6.44 · genuineness 3.91 · WER 0.9%
Fixed-target emotion MSE 2.467 (normalized 0.1542)

1.5 Alpha + real-speech SFT · rank 4

CE 6.48 · genuineness 4.07 · WER 0.0%
Fixed-target emotion MSE 2.229 (normalized 0.1393)

1.5 Alpha + real-speech SFT · rank 8

CE 6.55 · genuineness 4.03 · WER 0.9%
Fixed-target emotion MSE 2.771 (normalized 0.1732)

betrayal resolve · German · prompt_only

Ich habe heute Morgen die Ankündigung gesehen und dachte zuerst, es gäbe noch eine zweite Seite mit unseren anderen Namen. Aber da stand nur deiner, direkt unter der Idee, an der wir hier gemeinsam gearbeitet haben, als längst alle anderen zu Hause waren. Du hast sogar den kleinen Satz übernommen, den ich ans Fenster geschrieben habe... irgendwie ist das beinahe schon komisch. Ich verlange keine wunderschöne Entschuldigung; ich möchte, dass du die Ankündigung korrigierst und die Arbeit des ganzen Teams anerkennst. Danach können wir darüber reden, ob wir wieder zusammenarbeiten, aber erst musst du verstehen, warum ich das nicht einfach weglächeln kann.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Polite restraint fractures into hurt and bitterness, then becomes self-respecting resolve.
SCRIPT:
(carefully polite, hurt held beneath the surface) [6.4 seconds duration] Ich habe heute Morgen die Ankündigung gesehen und dachte zuerst, es gäbe noch eine zweite Seite mit unseren anderen Namen.
[0.3 seconds pause]
(disbelief, a little colder and slower) [6.5 seconds duration] Aber da stand nur deiner, direkt unter der Idee, an der wir hier gemeinsam gearbeitet haben, als längst alle anderen zu Hause waren.
[0.3 seconds pause]
(low bitter chuckle, 0.4 seconds)
(bitter and wounded, not villainous) [6.8 seconds duration] Du hast sogar den kleinen Satz übernommen, den ich ans Fenster geschrieben habe... irgendwie ist das beinahe schon komisch.
[0.3 seconds pause]
(firm, controlled anger, clearly articulated) [6.8 seconds duration] Ich verlange keine wunderschöne Entschuldigung; ich möchte, dass du die Ankündigung korrigierst und die Arbeit des ganzen Teams anerkennst.
[0.3 seconds pause]
(quiet dignity, a boundary stated without shouting) [6.7 seconds duration] Danach können wir darüber reden, ob wir wieder zusammenarbeiten, aber erst musst du verstehen, warum ich das nicht einfach weglächeln kann.

Fixed seed 20261526 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.38 · genuineness 4.38 · WER 0.0%
Fixed-target emotion MSE 2.455 (normalized 0.1534)

1.5 Alpha · merged, no additional SFT

CE 6.42 · genuineness 4.46 · WER 2.0%
Fixed-target emotion MSE 2.823 (normalized 0.1765)

1.5 Alpha + real-speech SFT · rank 1

CE 6.40 · genuineness 4.36 · WER 2.0%
Fixed-target emotion MSE 2.496 (normalized 0.1560)

1.5 Alpha + real-speech SFT · rank 2

CE 6.41 · genuineness 4.28 · WER 1.0%
Fixed-target emotion MSE 2.873 (normalized 0.1795)

1.5 Alpha + real-speech SFT · rank 4

CE 6.45 · genuineness 4.39 · WER 1.0%
Fixed-target emotion MSE 2.848 (normalized 0.1780)

1.5 Alpha + real-speech SFT · rank 8

CE 6.41 · genuineness 4.51 · WER 1.0%
Fixed-target emotion MSE 2.748 (normalized 0.1717)

betrayal resolve · German · reference

Ich habe heute Morgen die Ankündigung gesehen und dachte zuerst, es gäbe noch eine zweite Seite mit unseren anderen Namen. Aber da stand nur deiner, direkt unter der Idee, an der wir hier gemeinsam gearbeitet haben, als längst alle anderen zu Hause waren. Du hast sogar den kleinen Satz übernommen, den ich ans Fenster geschrieben habe... irgendwie ist das beinahe schon komisch. Ich verlange keine wunderschöne Entschuldigung; ich möchte, dass du die Ankündigung korrigierst und die Arbeit des ganzen Teams anerkennst. Danach können wir darüber reden, ob wir wieder zusammenarbeiten, aber erst musst du verstehen, warum ich das nicht einfach weglächeln kann.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Polite restraint fractures into hurt and bitterness, then becomes self-respecting resolve.
SCRIPT:
(carefully polite, hurt held beneath the surface) [6.4 seconds duration] Ich habe heute Morgen die Ankündigung gesehen und dachte zuerst, es gäbe noch eine zweite Seite mit unseren anderen Namen.
[0.3 seconds pause]
(disbelief, a little colder and slower) [6.5 seconds duration] Aber da stand nur deiner, direkt unter der Idee, an der wir hier gemeinsam gearbeitet haben, als längst alle anderen zu Hause waren.
[0.3 seconds pause]
(low bitter chuckle, 0.4 seconds)
(bitter and wounded, not villainous) [6.8 seconds duration] Du hast sogar den kleinen Satz übernommen, den ich ans Fenster geschrieben habe... irgendwie ist das beinahe schon komisch.
[0.3 seconds pause]
(firm, controlled anger, clearly articulated) [6.8 seconds duration] Ich verlange keine wunderschöne Entschuldigung; ich möchte, dass du die Ankündigung korrigierst und die Arbeit des ganzen Teams anerkennst.
[0.3 seconds pause]
(quiet dignity, a boundary stated without shouting) [6.7 seconds duration] Danach können wir darüber reden, ob wir wieder zusammenarbeiten, aber erst musst du verstehen, warum ich das nicht einfach weglächeln kann.

Fixed seed 20261526 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.46 · genuineness 4.50 · WER 1.0%
Fixed-target emotion MSE 2.324 (normalized 0.1453)

1.5 Alpha · merged, no additional SFT

CE 6.44 · genuineness 4.56 · WER 1.0%
Fixed-target emotion MSE 2.411 (normalized 0.1507)

1.5 Alpha + real-speech SFT · rank 1

CE 6.48 · genuineness 4.26 · WER 1.0%
Fixed-target emotion MSE 2.589 (normalized 0.1618)

1.5 Alpha + real-speech SFT · rank 2

CE 6.52 · genuineness 4.42 · WER 1.0%
Fixed-target emotion MSE 2.565 (normalized 0.1603)

1.5 Alpha + real-speech SFT · rank 4

CE 6.50 · genuineness 4.49 · WER 0.0%
Fixed-target emotion MSE 2.564 (normalized 0.1602)

1.5 Alpha + real-speech SFT · rank 8

CE 6.52 · genuineness 4.51 · WER 1.0%
Fixed-target emotion MSE 2.491 (normalized 0.1557)

betrayal resolve · Dutch · prompt_only

Ik zag de aankondiging vanochtend en dacht eerst dat er vast nog een tweede pagina met onze andere namen was. Maar alleen jouw naam stond er, onder het idee waar wij samen aan werkten terwijl iedereen al lang naar huis was. Je hebt zelfs dat zinnetje gebruikt dat ik op het raam schreef... op een heel ellendige manier is dat bijna grappig. Ik vraag niet om een prachtig excuus; ik wil dat je de aankondiging verbetert en het hele team de eer geeft. Daarna kunnen we praten over opnieuw samenwerken, maar eerst moet je begrijpen waarom ik dit niet zomaar kan weglachen.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Polite restraint fractures into hurt and bitterness, then becomes self-respecting resolve.
SCRIPT:
(carefully polite, hurt held beneath the surface) [6.4 seconds duration] Ik zag de aankondiging vanochtend en dacht eerst dat er vast nog een tweede pagina met onze andere namen was.
[0.3 seconds pause]
(disbelief, a little colder and slower) [6.5 seconds duration] Maar alleen jouw naam stond er, onder het idee waar wij samen aan werkten terwijl iedereen al lang naar huis was.
[0.3 seconds pause]
(low bitter chuckle, 0.4 seconds)
(bitter and wounded, not villainous) [6.8 seconds duration] Je hebt zelfs dat zinnetje gebruikt dat ik op het raam schreef... op een heel ellendige manier is dat bijna grappig.
[0.3 seconds pause]
(firm, controlled anger, clearly articulated) [6.8 seconds duration] Ik vraag niet om een prachtig excuus; ik wil dat je de aankondiging verbetert en het hele team de eer geeft.
[0.3 seconds pause]
(quiet dignity, a boundary stated without shouting) [6.7 seconds duration] Daarna kunnen we praten over opnieuw samenwerken, maar eerst moet je begrijpen waarom ik dit niet zomaar kan weglachen.

Fixed seed 20261526 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.27 · genuineness 4.36 · WER 0.0%
Fixed-target emotion MSE 2.817 (normalized 0.1760)

1.5 Alpha · merged, no additional SFT

CE 6.29 · genuineness 4.37 · WER 3.9%
Fixed-target emotion MSE 2.808 (normalized 0.1755)

1.5 Alpha + real-speech SFT · rank 1

CE 6.24 · genuineness 4.28 · WER 2.0%
Fixed-target emotion MSE 2.604 (normalized 0.1627)

1.5 Alpha + real-speech SFT · rank 2

CE 6.29 · genuineness 4.30 · WER 0.0%
Fixed-target emotion MSE 2.748 (normalized 0.1718)

1.5 Alpha + real-speech SFT · rank 4

CE 6.24 · genuineness 4.27 · WER 3.9%
Fixed-target emotion MSE 2.771 (normalized 0.1732)

1.5 Alpha + real-speech SFT · rank 8

CE 6.21 · genuineness 4.36 · WER 0.0%
Fixed-target emotion MSE 2.789 (normalized 0.1743)

betrayal resolve · Dutch · reference

Ik zag de aankondiging vanochtend en dacht eerst dat er vast nog een tweede pagina met onze andere namen was. Maar alleen jouw naam stond er, onder het idee waar wij samen aan werkten terwijl iedereen al lang naar huis was. Je hebt zelfs dat zinnetje gebruikt dat ik op het raam schreef... op een heel ellendige manier is dat bijna grappig. Ik vraag niet om een prachtig excuus; ik wil dat je de aankondiging verbetert en het hele team de eer geeft. Daarna kunnen we praten over opnieuw samenwerken, maar eerst moet je begrijpen waarom ik dit niet zomaar kan weglachen.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Polite restraint fractures into hurt and bitterness, then becomes self-respecting resolve.
SCRIPT:
(carefully polite, hurt held beneath the surface) [6.4 seconds duration] Ik zag de aankondiging vanochtend en dacht eerst dat er vast nog een tweede pagina met onze andere namen was.
[0.3 seconds pause]
(disbelief, a little colder and slower) [6.5 seconds duration] Maar alleen jouw naam stond er, onder het idee waar wij samen aan werkten terwijl iedereen al lang naar huis was.
[0.3 seconds pause]
(low bitter chuckle, 0.4 seconds)
(bitter and wounded, not villainous) [6.8 seconds duration] Je hebt zelfs dat zinnetje gebruikt dat ik op het raam schreef... op een heel ellendige manier is dat bijna grappig.
[0.3 seconds pause]
(firm, controlled anger, clearly articulated) [6.8 seconds duration] Ik vraag niet om een prachtig excuus; ik wil dat je de aankondiging verbetert en het hele team de eer geeft.
[0.3 seconds pause]
(quiet dignity, a boundary stated without shouting) [6.7 seconds duration] Daarna kunnen we praten over opnieuw samenwerken, maar eerst moet je begrijpen waarom ik dit niet zomaar kan weglachen.

Fixed seed 20261526 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.28 · genuineness 4.28 · WER 2.0%
Fixed-target emotion MSE 2.681 (normalized 0.1676)

1.5 Alpha · merged, no additional SFT

CE 6.15 · genuineness 4.28 · WER 0.0%
Fixed-target emotion MSE 2.674 (normalized 0.1671)

1.5 Alpha + real-speech SFT · rank 1

CE 6.27 · genuineness 4.22 · WER 0.0%
Fixed-target emotion MSE 2.853 (normalized 0.1783)

1.5 Alpha + real-speech SFT · rank 2

CE 6.28 · genuineness 4.03 · WER 0.0%
Fixed-target emotion MSE 2.706 (normalized 0.1692)

1.5 Alpha + real-speech SFT · rank 4

CE 6.28 · genuineness 4.34 · WER 4.9%
Fixed-target emotion MSE 2.654 (normalized 0.1659)

1.5 Alpha + real-speech SFT · rank 8

CE 6.29 · genuineness 4.15 · WER 3.9%
Fixed-target emotion MSE 2.831 (normalized 0.1769)

betrayal resolve · French · prompt_only

J'ai vu l'annonce ce matin, et j'ai d'abord pensé qu'il devait y avoir une autre page avec les noms de notre équipe. Mais il n'y avait que le tien, sous cette idée que nous avons construite ensemble quand tout le monde était déjà parti. Tu as même repris la petite phrase que j'avais écrite sur la fenêtre... c'est presque drôle, d'une façon vraiment triste. Je ne te demande pas de belles excuses; je te demande de corriger l'annonce et de reconnaître le travail de toute l'équipe. Ensuite, nous pourrons parler de travailler à nouveau ensemble, mais comprends d'abord pourquoi je ne peux pas simplement sourire et oublier.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Polite restraint fractures into hurt and bitterness, then becomes self-respecting resolve.
SCRIPT:
(carefully polite, hurt held beneath the surface) [6.4 seconds duration] J'ai vu l'annonce ce matin, et j'ai d'abord pensé qu'il devait y avoir une autre page avec les noms de notre équipe.
[0.3 seconds pause]
(disbelief, a little colder and slower) [6.5 seconds duration] Mais il n'y avait que le tien, sous cette idée que nous avons construite ensemble quand tout le monde était déjà parti.
[0.3 seconds pause]
(low bitter chuckle, 0.4 seconds)
(bitter and wounded, not villainous) [6.8 seconds duration] Tu as même repris la petite phrase que j'avais écrite sur la fenêtre... c'est presque drôle, d'une façon vraiment triste.
[0.3 seconds pause]
(firm, controlled anger, clearly articulated) [6.8 seconds duration] Je ne te demande pas de belles excuses; je te demande de corriger l'annonce et de reconnaître le travail de toute l'équipe.
[0.3 seconds pause]
(quiet dignity, a boundary stated without shouting) [6.7 seconds duration] Ensuite, nous pourrons parler de travailler à nouveau ensemble, mais comprends d'abord pourquoi je ne peux pas simplement sourire et oublier.

Fixed seed 20261526 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.15 · genuineness 4.56 · WER 0.0%
Fixed-target emotion MSE 2.695 (normalized 0.1684)

1.5 Alpha · merged, no additional SFT

CE 6.14 · genuineness 4.42 · WER 0.9%
Fixed-target emotion MSE 2.623 (normalized 0.1639)

1.5 Alpha + real-speech SFT · rank 1

CE 6.11 · genuineness 4.44 · WER 0.0%
Fixed-target emotion MSE 2.653 (normalized 0.1658)

1.5 Alpha + real-speech SFT · rank 2

CE 6.11 · genuineness 4.41 · WER 0.0%
Fixed-target emotion MSE 2.752 (normalized 0.1720)

1.5 Alpha + real-speech SFT · rank 4

CE 6.03 · genuineness 4.53 · WER 0.0%
Fixed-target emotion MSE 2.585 (normalized 0.1615)

1.5 Alpha + real-speech SFT · rank 8

CE 6.22 · genuineness 4.51 · WER 0.0%
Fixed-target emotion MSE 2.758 (normalized 0.1724)

betrayal resolve · French · reference

J'ai vu l'annonce ce matin, et j'ai d'abord pensé qu'il devait y avoir une autre page avec les noms de notre équipe. Mais il n'y avait que le tien, sous cette idée que nous avons construite ensemble quand tout le monde était déjà parti. Tu as même repris la petite phrase que j'avais écrite sur la fenêtre... c'est presque drôle, d'une façon vraiment triste. Je ne te demande pas de belles excuses; je te demande de corriger l'annonce et de reconnaître le travail de toute l'équipe. Ensuite, nous pourrons parler de travailler à nouveau ensemble, mais comprends d'abord pourquoi je ne peux pas simplement sourire et oublier.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Polite restraint fractures into hurt and bitterness, then becomes self-respecting resolve.
SCRIPT:
(carefully polite, hurt held beneath the surface) [6.4 seconds duration] J'ai vu l'annonce ce matin, et j'ai d'abord pensé qu'il devait y avoir une autre page avec les noms de notre équipe.
[0.3 seconds pause]
(disbelief, a little colder and slower) [6.5 seconds duration] Mais il n'y avait que le tien, sous cette idée que nous avons construite ensemble quand tout le monde était déjà parti.
[0.3 seconds pause]
(low bitter chuckle, 0.4 seconds)
(bitter and wounded, not villainous) [6.8 seconds duration] Tu as même repris la petite phrase que j'avais écrite sur la fenêtre... c'est presque drôle, d'une façon vraiment triste.
[0.3 seconds pause]
(firm, controlled anger, clearly articulated) [6.8 seconds duration] Je ne te demande pas de belles excuses; je te demande de corriger l'annonce et de reconnaître le travail de toute l'équipe.
[0.3 seconds pause]
(quiet dignity, a boundary stated without shouting) [6.7 seconds duration] Ensuite, nous pourrons parler de travailler à nouveau ensemble, mais comprends d'abord pourquoi je ne peux pas simplement sourire et oublier.

Fixed seed 20261526 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.27 · genuineness 4.36 · WER 0.0%
Fixed-target emotion MSE 2.726 (normalized 0.1703)

1.5 Alpha · merged, no additional SFT

CE 6.07 · genuineness 4.56 · WER 0.0%
Fixed-target emotion MSE 2.088 (normalized 0.1305)

1.5 Alpha + real-speech SFT · rank 1

CE 6.22 · genuineness 4.41 · WER 0.0%
Fixed-target emotion MSE 2.654 (normalized 0.1659)

1.5 Alpha + real-speech SFT · rank 2

CE 6.09 · genuineness 4.49 · WER 0.0%
Fixed-target emotion MSE 2.597 (normalized 0.1623)

1.5 Alpha + real-speech SFT · rank 4

CE 6.22 · genuineness 4.46 · WER 0.0%
Fixed-target emotion MSE 2.252 (normalized 0.1407)

1.5 Alpha + real-speech SFT · rank 8

CE 6.17 · genuineness 4.40 · WER 0.0%
Fixed-target emotion MSE 2.387 (normalized 0.1492)

betrayal resolve · Spanish · prompt_only

Vi el anuncio esta mañana y al principio pensé que habría otra página con los nombres de las demás personas del equipo. Pero solamente estaba el tuyo, debajo de la idea que construimos juntos cuando todos los demás ya se habían marchado a casa. Hasta utilizaste aquella frase que escribí en la ventana... casi tiene gracia, pero de una manera bastante triste, la verdad. No te estoy pidiendo una disculpa preciosa; te estoy pidiendo que corrijas el anuncio y reconozcas el trabajo de todo el equipo. Después podremos hablar de volver a trabajar juntos, pero primero tienes que entender por qué no puedo sonreír y olvidarlo sin más.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Polite restraint fractures into hurt and bitterness, then becomes self-respecting resolve.
SCRIPT:
(carefully polite, hurt held beneath the surface) [6.4 seconds duration] Vi el anuncio esta mañana y al principio pensé que habría otra página con los nombres de las demás personas del equipo.
[0.3 seconds pause]
(disbelief, a little colder and slower) [6.5 seconds duration] Pero solamente estaba el tuyo, debajo de la idea que construimos juntos cuando todos los demás ya se habían marchado a casa.
[0.3 seconds pause]
(low bitter chuckle, 0.4 seconds)
(bitter and wounded, not villainous) [6.8 seconds duration] Hasta utilizaste aquella frase que escribí en la ventana... casi tiene gracia, pero de una manera bastante triste, la verdad.
[0.3 seconds pause]
(firm, controlled anger, clearly articulated) [6.8 seconds duration] No te estoy pidiendo una disculpa preciosa; te estoy pidiendo que corrijas el anuncio y reconozcas el trabajo de todo el equipo.
[0.3 seconds pause]
(quiet dignity, a boundary stated without shouting) [6.7 seconds duration] Después podremos hablar de volver a trabajar juntos, pero primero tienes que entender por qué no puedo sonreír y olvidarlo sin más.

Fixed seed 20261526 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.32 · genuineness 4.41 · WER 0.0%
Fixed-target emotion MSE 0.806 (normalized 0.0504)

1.5 Alpha · merged, no additional SFT

CE 6.24 · genuineness 4.41 · WER 0.9%
Fixed-target emotion MSE 1.523 (normalized 0.0952)

1.5 Alpha + real-speech SFT · rank 1

CE 6.24 · genuineness 4.41 · WER 0.0%
Fixed-target emotion MSE 0.849 (normalized 0.0530)

1.5 Alpha + real-speech SFT · rank 2

CE 6.28 · genuineness 4.34 · WER 0.9%
Fixed-target emotion MSE 1.848 (normalized 0.1155)

1.5 Alpha + real-speech SFT · rank 4

CE 6.35 · genuineness 4.43 · WER 0.0%
Fixed-target emotion MSE 1.810 (normalized 0.1131)

1.5 Alpha + real-speech SFT · rank 8

CE 6.25 · genuineness 4.43 · WER 0.0%
Fixed-target emotion MSE 1.907 (normalized 0.1192)

betrayal resolve · Spanish · reference

Vi el anuncio esta mañana y al principio pensé que habría otra página con los nombres de las demás personas del equipo. Pero solamente estaba el tuyo, debajo de la idea que construimos juntos cuando todos los demás ya se habían marchado a casa. Hasta utilizaste aquella frase que escribí en la ventana... casi tiene gracia, pero de una manera bastante triste, la verdad. No te estoy pidiendo una disculpa preciosa; te estoy pidiendo que corrijas el anuncio y reconozcas el trabajo de todo el equipo. Después podremos hablar de volver a trabajar juntos, pero primero tienes que entender por qué no puedo sonreír y olvidarlo sin más.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Polite restraint fractures into hurt and bitterness, then becomes self-respecting resolve.
SCRIPT:
(carefully polite, hurt held beneath the surface) [6.4 seconds duration] Vi el anuncio esta mañana y al principio pensé que habría otra página con los nombres de las demás personas del equipo.
[0.3 seconds pause]
(disbelief, a little colder and slower) [6.5 seconds duration] Pero solamente estaba el tuyo, debajo de la idea que construimos juntos cuando todos los demás ya se habían marchado a casa.
[0.3 seconds pause]
(low bitter chuckle, 0.4 seconds)
(bitter and wounded, not villainous) [6.8 seconds duration] Hasta utilizaste aquella frase que escribí en la ventana... casi tiene gracia, pero de una manera bastante triste, la verdad.
[0.3 seconds pause]
(firm, controlled anger, clearly articulated) [6.8 seconds duration] No te estoy pidiendo una disculpa preciosa; te estoy pidiendo que corrijas el anuncio y reconozcas el trabajo de todo el equipo.
[0.3 seconds pause]
(quiet dignity, a boundary stated without shouting) [6.7 seconds duration] Después podremos hablar de volver a trabajar juntos, pero primero tienes que entender por qué no puedo sonreír y olvidarlo sin más.

Fixed seed 20261526 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.12 · genuineness 4.53 · WER 0.0%
Fixed-target emotion MSE 2.333 (normalized 0.1458)

1.5 Alpha · merged, no additional SFT

CE 6.23 · genuineness 4.05 · WER 0.9%
Fixed-target emotion MSE 2.469 (normalized 0.1543)

1.5 Alpha + real-speech SFT · rank 1

CE 6.22 · genuineness 4.12 · WER 0.9%
Fixed-target emotion MSE 2.037 (normalized 0.1273)

1.5 Alpha + real-speech SFT · rank 2

CE 6.13 · genuineness 4.19 · WER 0.0%
Fixed-target emotion MSE 2.051 (normalized 0.1282)

1.5 Alpha + real-speech SFT · rank 4

CE 6.16 · genuineness 4.06 · WER 0.0%
Fixed-target emotion MSE 2.241 (normalized 0.1401)

1.5 Alpha + real-speech SFT · rank 8

CE 6.26 · genuineness 4.12 · WER 0.0%
Fixed-target emotion MSE 2.431 (normalized 0.1519)

awe tenderness · English · prompt_only

Look at the edge of the clouds, just there; I thought the sunrise would be bright, but I did not expect that color. All those months of planning, and now the whole world seems to be waking up beneath this little window beside us. I keep trying to think of something clever to say... but honestly, I feel very small, and a little overwhelmed. You do not have to say anything either; put your hand here and let us just watch it change for a moment. When we get home, I will forget the exact shape of the clouds, but I do not think I will forget being here with you.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Breathless awe, a little vulnerability and quiet, shared tenderness.
SCRIPT:
(breathless wonder, almost a whisper) [6.4 seconds duration] Look at the edge of the clouds, just there; I thought the sunrise would be bright, but I did not expect that color.
[0.3 seconds pause]
(expansive awe, gradually opening the voice) [6.5 seconds duration] All those months of planning, and now the whole world seems to be waking up beneath this little window beside us.
[0.3 seconds pause]
(soft amazed exhale, 0.4 seconds)
(vulnerable honesty, a tiny searching pause) [6.8 seconds duration] I keep trying to think of something clever to say... but honestly, I feel very small, and a little overwhelmed.
[0.3 seconds pause]
(soft affection and reassurance) [6.8 seconds duration] You do not have to say anything either; put your hand here and let us just watch it change for a moment.
[0.3 seconds pause]
(quiet inspiration, intimate rather than grandiose) [6.7 seconds duration] When we get home, I will forget the exact shape of the clouds, but I do not think I will forget being here with you.

Fixed seed 20261627 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.45 · genuineness 4.26 · WER 0.0%
Fixed-target emotion MSE 1.066 (normalized 0.0666)

1.5 Alpha · merged, no additional SFT

CE 6.48 · genuineness 4.17 · WER 0.0%
Fixed-target emotion MSE 0.569 (normalized 0.0356)

1.5 Alpha + real-speech SFT · rank 1

CE 6.59 · genuineness 4.47 · WER 0.0%
Fixed-target emotion MSE 0.661 (normalized 0.0413)

1.5 Alpha + real-speech SFT · rank 2

CE 6.52 · genuineness 4.39 · WER 0.0%
Fixed-target emotion MSE 0.578 (normalized 0.0361)

1.5 Alpha + real-speech SFT · rank 4

CE 6.55 · genuineness 4.49 · WER 0.0%
Fixed-target emotion MSE 0.464 (normalized 0.0290)

1.5 Alpha + real-speech SFT · rank 8

CE 6.55 · genuineness 4.49 · WER 0.0%
Fixed-target emotion MSE 0.591 (normalized 0.0370)

awe tenderness · English · reference

Look at the edge of the clouds, just there; I thought the sunrise would be bright, but I did not expect that color. All those months of planning, and now the whole world seems to be waking up beneath this little window beside us. I keep trying to think of something clever to say... but honestly, I feel very small, and a little overwhelmed. You do not have to say anything either; put your hand here and let us just watch it change for a moment. When we get home, I will forget the exact shape of the clouds, but I do not think I will forget being here with you.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Breathless awe, a little vulnerability and quiet, shared tenderness.
SCRIPT:
(breathless wonder, almost a whisper) [6.4 seconds duration] Look at the edge of the clouds, just there; I thought the sunrise would be bright, but I did not expect that color.
[0.3 seconds pause]
(expansive awe, gradually opening the voice) [6.5 seconds duration] All those months of planning, and now the whole world seems to be waking up beneath this little window beside us.
[0.3 seconds pause]
(soft amazed exhale, 0.4 seconds)
(vulnerable honesty, a tiny searching pause) [6.8 seconds duration] I keep trying to think of something clever to say... but honestly, I feel very small, and a little overwhelmed.
[0.3 seconds pause]
(soft affection and reassurance) [6.8 seconds duration] You do not have to say anything either; put your hand here and let us just watch it change for a moment.
[0.3 seconds pause]
(quiet inspiration, intimate rather than grandiose) [6.7 seconds duration] When we get home, I will forget the exact shape of the clouds, but I do not think I will forget being here with you.

Fixed seed 20261627 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.57 · genuineness 4.43 · WER 0.0%
Fixed-target emotion MSE 1.003 (normalized 0.0627)

1.5 Alpha · merged, no additional SFT

CE 6.54 · genuineness 4.25 · WER 0.0%
Fixed-target emotion MSE 0.737 (normalized 0.0461)

1.5 Alpha + real-speech SFT · rank 1

CE 6.50 · genuineness 4.40 · WER 0.0%
Fixed-target emotion MSE 0.685 (normalized 0.0428)

1.5 Alpha + real-speech SFT · rank 2

CE 6.58 · genuineness 4.44 · WER 0.0%
Fixed-target emotion MSE 0.632 (normalized 0.0395)

1.5 Alpha + real-speech SFT · rank 4

CE 6.46 · genuineness 4.36 · WER 0.0%
Fixed-target emotion MSE 0.580 (normalized 0.0363)

1.5 Alpha + real-speech SFT · rank 8

CE 6.56 · genuineness 4.38 · WER 0.0%
Fixed-target emotion MSE 0.724 (normalized 0.0452)

awe tenderness · German · prompt_only

Schau mal an den Rand der Wolken, genau dort; ich wusste, der Sonnenaufgang wird hell, aber mit dieser Farbe habe ich nicht gerechnet. Nach all diesen Monaten der Planung scheint jetzt die ganze Welt unter unserem kleinen Fenster langsam wieder aufzuwachen. Ich versuche die ganze Zeit, etwas Kluges zu sagen... aber ehrlich gesagt fühle ich mich gerade sehr klein und ziemlich überwältigt. Du musst auch nichts sagen; leg einfach deine Hand hierher, und lass uns einen Moment gemeinsam zusehen, wie sich alles verändert. Zu Hause werde ich die genaue Form der Wolken vergessen, aber nicht, wie es war, diesen Augenblick mit dir zu erleben.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Breathless awe, a little vulnerability and quiet, shared tenderness.
SCRIPT:
(breathless wonder, almost a whisper) [6.4 seconds duration] Schau mal an den Rand der Wolken, genau dort; ich wusste, der Sonnenaufgang wird hell, aber mit dieser Farbe habe ich nicht gerechnet.
[0.3 seconds pause]
(expansive awe, gradually opening the voice) [6.5 seconds duration] Nach all diesen Monaten der Planung scheint jetzt die ganze Welt unter unserem kleinen Fenster langsam wieder aufzuwachen.
[0.3 seconds pause]
(soft amazed exhale, 0.4 seconds)
(vulnerable honesty, a tiny searching pause) [6.8 seconds duration] Ich versuche die ganze Zeit, etwas Kluges zu sagen... aber ehrlich gesagt fühle ich mich gerade sehr klein und ziemlich überwältigt.
[0.3 seconds pause]
(soft affection and reassurance) [6.8 seconds duration] Du musst auch nichts sagen; leg einfach deine Hand hierher, und lass uns einen Moment gemeinsam zusehen, wie sich alles verändert.
[0.3 seconds pause]
(quiet inspiration, intimate rather than grandiose) [6.7 seconds duration] Zu Hause werde ich die genaue Form der Wolken vergessen, aber nicht, wie es war, diesen Augenblick mit dir zu erleben.

Fixed seed 20261627 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.48 · genuineness 4.61 · WER 0.0%
Fixed-target emotion MSE 1.478 (normalized 0.0924)

1.5 Alpha · merged, no additional SFT

CE 6.42 · genuineness 4.61 · WER 1.9%
Fixed-target emotion MSE 1.439 (normalized 0.0899)

1.5 Alpha + real-speech SFT · rank 1

CE 6.35 · genuineness 4.51 · WER 0.0%
Fixed-target emotion MSE 1.757 (normalized 0.1098)

1.5 Alpha + real-speech SFT · rank 2

CE 6.44 · genuineness 4.47 · WER 0.0%
Fixed-target emotion MSE 1.196 (normalized 0.0748)

1.5 Alpha + real-speech SFT · rank 4

CE 6.40 · genuineness 4.67 · WER 0.0%
Fixed-target emotion MSE 0.845 (normalized 0.0528)

1.5 Alpha + real-speech SFT · rank 8

CE 6.45 · genuineness 4.69 · WER 0.0%
Fixed-target emotion MSE 1.111 (normalized 0.0695)

awe tenderness · German · reference

Schau mal an den Rand der Wolken, genau dort; ich wusste, der Sonnenaufgang wird hell, aber mit dieser Farbe habe ich nicht gerechnet. Nach all diesen Monaten der Planung scheint jetzt die ganze Welt unter unserem kleinen Fenster langsam wieder aufzuwachen. Ich versuche die ganze Zeit, etwas Kluges zu sagen... aber ehrlich gesagt fühle ich mich gerade sehr klein und ziemlich überwältigt. Du musst auch nichts sagen; leg einfach deine Hand hierher, und lass uns einen Moment gemeinsam zusehen, wie sich alles verändert. Zu Hause werde ich die genaue Form der Wolken vergessen, aber nicht, wie es war, diesen Augenblick mit dir zu erleben.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Breathless awe, a little vulnerability and quiet, shared tenderness.
SCRIPT:
(breathless wonder, almost a whisper) [6.4 seconds duration] Schau mal an den Rand der Wolken, genau dort; ich wusste, der Sonnenaufgang wird hell, aber mit dieser Farbe habe ich nicht gerechnet.
[0.3 seconds pause]
(expansive awe, gradually opening the voice) [6.5 seconds duration] Nach all diesen Monaten der Planung scheint jetzt die ganze Welt unter unserem kleinen Fenster langsam wieder aufzuwachen.
[0.3 seconds pause]
(soft amazed exhale, 0.4 seconds)
(vulnerable honesty, a tiny searching pause) [6.8 seconds duration] Ich versuche die ganze Zeit, etwas Kluges zu sagen... aber ehrlich gesagt fühle ich mich gerade sehr klein und ziemlich überwältigt.
[0.3 seconds pause]
(soft affection and reassurance) [6.8 seconds duration] Du musst auch nichts sagen; leg einfach deine Hand hierher, und lass uns einen Moment gemeinsam zusehen, wie sich alles verändert.
[0.3 seconds pause]
(quiet inspiration, intimate rather than grandiose) [6.7 seconds duration] Zu Hause werde ich die genaue Form der Wolken vergessen, aber nicht, wie es war, diesen Augenblick mit dir zu erleben.

Fixed seed 20261627 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.42 · genuineness 4.72 · WER 0.0%
Fixed-target emotion MSE 1.159 (normalized 0.0724)

1.5 Alpha · merged, no additional SFT

CE 6.46 · genuineness 4.61 · WER 1.0%
Fixed-target emotion MSE 0.745 (normalized 0.0465)

1.5 Alpha + real-speech SFT · rank 1

CE 6.46 · genuineness 4.64 · WER 1.0%
Fixed-target emotion MSE 1.059 (normalized 0.0662)

1.5 Alpha + real-speech SFT · rank 2

CE 6.48 · genuineness 4.67 · WER 0.0%
Fixed-target emotion MSE 1.039 (normalized 0.0650)

1.5 Alpha + real-speech SFT · rank 4

CE 6.33 · genuineness 4.62 · WER 1.0%
Fixed-target emotion MSE 1.047 (normalized 0.0654)

1.5 Alpha + real-speech SFT · rank 8

CE 6.47 · genuineness 4.68 · WER 0.0%
Fixed-target emotion MSE 0.806 (normalized 0.0504)

awe tenderness · Dutch · prompt_only

Kijk naar de rand van de wolken, precies daar; ik wist dat de zonsopgang fel zou zijn, maar die kleur had ik niet verwacht. Na al die maanden plannen lijkt nu de hele wereld langzaam wakker te worden onder dit kleine raam naast ons. Ik probeer steeds iets verstandigs te bedenken om te zeggen... maar eerlijk gezegd voel ik me heel klein en een beetje overweldigd. Jij hoeft ook niets te zeggen; leg je hand maar hier en laten we gewoon even samen kijken hoe alles verandert. Thuis zal ik de precieze vorm van de wolken vergeten, maar volgens mij vergeet ik nooit hoe het was om hier met jou te zijn.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Breathless awe, a little vulnerability and quiet, shared tenderness.
SCRIPT:
(breathless wonder, almost a whisper) [6.4 seconds duration] Kijk naar de rand van de wolken, precies daar; ik wist dat de zonsopgang fel zou zijn, maar die kleur had ik niet verwacht.
[0.3 seconds pause]
(expansive awe, gradually opening the voice) [6.5 seconds duration] Na al die maanden plannen lijkt nu de hele wereld langzaam wakker te worden onder dit kleine raam naast ons.
[0.3 seconds pause]
(soft amazed exhale, 0.4 seconds)
(vulnerable honesty, a tiny searching pause) [6.8 seconds duration] Ik probeer steeds iets verstandigs te bedenken om te zeggen... maar eerlijk gezegd voel ik me heel klein en een beetje overweldigd.
[0.3 seconds pause]
(soft affection and reassurance) [6.8 seconds duration] Jij hoeft ook niets te zeggen; leg je hand maar hier en laten we gewoon even samen kijken hoe alles verandert.
[0.3 seconds pause]
(quiet inspiration, intimate rather than grandiose) [6.7 seconds duration] Thuis zal ik de precieze vorm van de wolken vergeten, maar volgens mij vergeet ik nooit hoe het was om hier met jou te zijn.

Fixed seed 20261627 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.28 · genuineness 4.17 · WER 0.0%
Fixed-target emotion MSE 3.312 (normalized 0.2070)

1.5 Alpha · merged, no additional SFT

CE 6.34 · genuineness 4.32 · WER 0.0%
Fixed-target emotion MSE 3.241 (normalized 0.2026)

1.5 Alpha + real-speech SFT · rank 1

CE 6.29 · genuineness 4.17 · WER 0.9%
Fixed-target emotion MSE 3.233 (normalized 0.2020)

1.5 Alpha + real-speech SFT · rank 2

CE 6.25 · genuineness 4.30 · WER 0.0%
Fixed-target emotion MSE 4.183 (normalized 0.2614)

1.5 Alpha + real-speech SFT · rank 4

CE 6.34 · genuineness 4.33 · WER 0.0%
Fixed-target emotion MSE 3.375 (normalized 0.2110)

1.5 Alpha + real-speech SFT · rank 8

CE 6.27 · genuineness 4.36 · WER 1.8%
Fixed-target emotion MSE 2.886 (normalized 0.1804)

awe tenderness · Dutch · reference

Kijk naar de rand van de wolken, precies daar; ik wist dat de zonsopgang fel zou zijn, maar die kleur had ik niet verwacht. Na al die maanden plannen lijkt nu de hele wereld langzaam wakker te worden onder dit kleine raam naast ons. Ik probeer steeds iets verstandigs te bedenken om te zeggen... maar eerlijk gezegd voel ik me heel klein en een beetje overweldigd. Jij hoeft ook niets te zeggen; leg je hand maar hier en laten we gewoon even samen kijken hoe alles verandert. Thuis zal ik de precieze vorm van de wolken vergeten, maar volgens mij vergeet ik nooit hoe het was om hier met jou te zijn.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Breathless awe, a little vulnerability and quiet, shared tenderness.
SCRIPT:
(breathless wonder, almost a whisper) [6.4 seconds duration] Kijk naar de rand van de wolken, precies daar; ik wist dat de zonsopgang fel zou zijn, maar die kleur had ik niet verwacht.
[0.3 seconds pause]
(expansive awe, gradually opening the voice) [6.5 seconds duration] Na al die maanden plannen lijkt nu de hele wereld langzaam wakker te worden onder dit kleine raam naast ons.
[0.3 seconds pause]
(soft amazed exhale, 0.4 seconds)
(vulnerable honesty, a tiny searching pause) [6.8 seconds duration] Ik probeer steeds iets verstandigs te bedenken om te zeggen... maar eerlijk gezegd voel ik me heel klein en een beetje overweldigd.
[0.3 seconds pause]
(soft affection and reassurance) [6.8 seconds duration] Jij hoeft ook niets te zeggen; leg je hand maar hier en laten we gewoon even samen kijken hoe alles verandert.
[0.3 seconds pause]
(quiet inspiration, intimate rather than grandiose) [6.7 seconds duration] Thuis zal ik de precieze vorm van de wolken vergeten, maar volgens mij vergeet ik nooit hoe het was om hier met jou te zijn.

Fixed seed 20261627 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.30 · genuineness 4.37 · WER 0.9%
Fixed-target emotion MSE 3.530 (normalized 0.2206)

1.5 Alpha · merged, no additional SFT

CE 6.31 · genuineness 4.26 · WER 0.9%
Fixed-target emotion MSE 4.069 (normalized 0.2543)

1.5 Alpha + real-speech SFT · rank 1

CE 6.27 · genuineness 4.24 · WER 0.9%
Fixed-target emotion MSE 3.202 (normalized 0.2002)

1.5 Alpha + real-speech SFT · rank 2

CE 6.29 · genuineness 4.23 · WER 0.0%
Fixed-target emotion MSE 3.255 (normalized 0.2035)

1.5 Alpha + real-speech SFT · rank 4

CE 6.24 · genuineness 4.17 · WER 0.0%
Fixed-target emotion MSE 3.796 (normalized 0.2373)

1.5 Alpha + real-speech SFT · rank 8

CE 6.33 · genuineness 4.16 · WER 1.8%
Fixed-target emotion MSE 3.149 (normalized 0.1968)

awe tenderness · French · prompt_only

Regarde le bord des nuages, juste là; je savais que le lever du soleil serait lumineux, mais je n'imaginais pas cette couleur. Après tous ces mois de préparation, on dirait que le monde entier se réveille doucement sous cette petite fenêtre près de nous. Je cherche quelque chose d'intelligent à dire... mais en vérité, je me sens tout petit et un peu dépassé par ce moment. Tu n'as pas besoin de parler non plus; pose ta main ici et regardons simplement la lumière changer pendant quelques instants. En rentrant, j'oublierai sans doute la forme exacte des nuages, mais je crois que je n'oublierai jamais d'avoir été ici avec toi.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Breathless awe, a little vulnerability and quiet, shared tenderness.
SCRIPT:
(breathless wonder, almost a whisper) [6.4 seconds duration] Regarde le bord des nuages, juste là; je savais que le lever du soleil serait lumineux, mais je n'imaginais pas cette couleur.
[0.3 seconds pause]
(expansive awe, gradually opening the voice) [6.5 seconds duration] Après tous ces mois de préparation, on dirait que le monde entier se réveille doucement sous cette petite fenêtre près de nous.
[0.3 seconds pause]
(soft amazed exhale, 0.4 seconds)
(vulnerable honesty, a tiny searching pause) [6.8 seconds duration] Je cherche quelque chose d'intelligent à dire... mais en vérité, je me sens tout petit et un peu dépassé par ce moment.
[0.3 seconds pause]
(soft affection and reassurance) [6.8 seconds duration] Tu n'as pas besoin de parler non plus; pose ta main ici et regardons simplement la lumière changer pendant quelques instants.
[0.3 seconds pause]
(quiet inspiration, intimate rather than grandiose) [6.7 seconds duration] En rentrant, j'oublierai sans doute la forme exacte des nuages, mais je crois que je n'oublierai jamais d'avoir été ici avec toi.

Fixed seed 20261627 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.27 · genuineness 4.60 · WER 2.8%
Fixed-target emotion MSE 1.560 (normalized 0.0975)

1.5 Alpha · merged, no additional SFT

CE 6.22 · genuineness 4.61 · WER 1.8%
Fixed-target emotion MSE 1.887 (normalized 0.1180)

1.5 Alpha + real-speech SFT · rank 1

CE 6.23 · genuineness 4.49 · WER 1.8%
Fixed-target emotion MSE 1.545 (normalized 0.0965)

1.5 Alpha + real-speech SFT · rank 2

CE 6.22 · genuineness 4.58 · WER 2.8%
Fixed-target emotion MSE 1.256 (normalized 0.0785)

1.5 Alpha + real-speech SFT · rank 4

CE 6.34 · genuineness 4.61 · WER 1.8%
Fixed-target emotion MSE 2.206 (normalized 0.1379)

1.5 Alpha + real-speech SFT · rank 8

CE 6.23 · genuineness 4.47 · WER 8.3%
Fixed-target emotion MSE 1.648 (normalized 0.1030)

awe tenderness · French · reference

Regarde le bord des nuages, juste là; je savais que le lever du soleil serait lumineux, mais je n'imaginais pas cette couleur. Après tous ces mois de préparation, on dirait que le monde entier se réveille doucement sous cette petite fenêtre près de nous. Je cherche quelque chose d'intelligent à dire... mais en vérité, je me sens tout petit et un peu dépassé par ce moment. Tu n'as pas besoin de parler non plus; pose ta main ici et regardons simplement la lumière changer pendant quelques instants. En rentrant, j'oublierai sans doute la forme exacte des nuages, mais je crois que je n'oublierai jamais d'avoir été ici avec toi.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Breathless awe, a little vulnerability and quiet, shared tenderness.
SCRIPT:
(breathless wonder, almost a whisper) [6.4 seconds duration] Regarde le bord des nuages, juste là; je savais que le lever du soleil serait lumineux, mais je n'imaginais pas cette couleur.
[0.3 seconds pause]
(expansive awe, gradually opening the voice) [6.5 seconds duration] Après tous ces mois de préparation, on dirait que le monde entier se réveille doucement sous cette petite fenêtre près de nous.
[0.3 seconds pause]
(soft amazed exhale, 0.4 seconds)
(vulnerable honesty, a tiny searching pause) [6.8 seconds duration] Je cherche quelque chose d'intelligent à dire... mais en vérité, je me sens tout petit et un peu dépassé par ce moment.
[0.3 seconds pause]
(soft affection and reassurance) [6.8 seconds duration] Tu n'as pas besoin de parler non plus; pose ta main ici et regardons simplement la lumière changer pendant quelques instants.
[0.3 seconds pause]
(quiet inspiration, intimate rather than grandiose) [6.7 seconds duration] En rentrant, j'oublierai sans doute la forme exacte des nuages, mais je crois que je n'oublierai jamais d'avoir été ici avec toi.

Fixed seed 20261627 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.23 · genuineness 4.58 · WER 0.9%
Fixed-target emotion MSE 2.155 (normalized 0.1347)

1.5 Alpha · merged, no additional SFT

CE 6.25 · genuineness 4.66 · WER 0.9%
Fixed-target emotion MSE 2.502 (normalized 0.1564)

1.5 Alpha + real-speech SFT · rank 1

CE 6.28 · genuineness 4.71 · WER 0.9%
Fixed-target emotion MSE 2.448 (normalized 0.1530)

1.5 Alpha + real-speech SFT · rank 2

CE 6.21 · genuineness 4.62 · WER 0.0%
Fixed-target emotion MSE 2.621 (normalized 0.1638)

1.5 Alpha + real-speech SFT · rank 4

CE 6.27 · genuineness 4.70 · WER 0.0%
Fixed-target emotion MSE 2.141 (normalized 0.1338)

1.5 Alpha + real-speech SFT · rank 8

CE 6.23 · genuineness 4.55 · WER 0.9%
Fixed-target emotion MSE 2.261 (normalized 0.1413)

awe tenderness · Spanish · prompt_only

Mira el borde de las nubes, justo ahí; sabía que el amanecer sería luminoso, pero no me esperaba ese color. Después de tantos meses de preparativos, parece que el mundo entero está despertando despacio debajo de esta pequeña ventana junto a nosotros. Llevo un rato intentando pensar en algo inteligente que decir... pero sinceramente, me siento muy pequeño y un poco sobrepasado. Tú tampoco tienes que decir nada; pon la mano aquí y vamos a mirar juntos cómo cambia la luz durante un momento. Cuando volvamos a casa olvidaré la forma exacta de las nubes, pero creo que nunca olvidaré haber compartido este instante contigo.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Breathless awe, a little vulnerability and quiet, shared tenderness.
SCRIPT:
(breathless wonder, almost a whisper) [6.4 seconds duration] Mira el borde de las nubes, justo ahí; sabía que el amanecer sería luminoso, pero no me esperaba ese color.
[0.3 seconds pause]
(expansive awe, gradually opening the voice) [6.5 seconds duration] Después de tantos meses de preparativos, parece que el mundo entero está despertando despacio debajo de esta pequeña ventana junto a nosotros.
[0.3 seconds pause]
(soft amazed exhale, 0.4 seconds)
(vulnerable honesty, a tiny searching pause) [6.8 seconds duration] Llevo un rato intentando pensar en algo inteligente que decir... pero sinceramente, me siento muy pequeño y un poco sobrepasado.
[0.3 seconds pause]
(soft affection and reassurance) [6.8 seconds duration] Tú tampoco tienes que decir nada; pon la mano aquí y vamos a mirar juntos cómo cambia la luz durante un momento.
[0.3 seconds pause]
(quiet inspiration, intimate rather than grandiose) [6.7 seconds duration] Cuando volvamos a casa olvidaré la forma exacta de las nubes, pero creo que nunca olvidaré haber compartido este instante contigo.

Fixed seed 20261627 · requested 435 frames · no reference

Published 1.5 Alpha · unmerged original

CE 6.29 · genuineness 4.33 · WER 1.0%
Fixed-target emotion MSE 2.671 (normalized 0.1670)

1.5 Alpha · merged, no additional SFT

CE 6.25 · genuineness 4.23 · WER 0.0%
Fixed-target emotion MSE 2.176 (normalized 0.1360)

1.5 Alpha + real-speech SFT · rank 1

CE 6.19 · genuineness 4.33 · WER 0.0%
Fixed-target emotion MSE 2.980 (normalized 0.1862)

1.5 Alpha + real-speech SFT · rank 2

CE 6.25 · genuineness 4.38 · WER 1.0%
Fixed-target emotion MSE 2.557 (normalized 0.1598)

1.5 Alpha + real-speech SFT · rank 4

CE 6.09 · genuineness 4.22 · WER 0.0%
Fixed-target emotion MSE 3.420 (normalized 0.2138)

1.5 Alpha + real-speech SFT · rank 8

CE 6.15 · genuineness 4.22 · WER 0.0%
Fixed-target emotion MSE 2.921 (normalized 0.1826)

awe tenderness · Spanish · reference

Mira el borde de las nubes, justo ahí; sabía que el amanecer sería luminoso, pero no me esperaba ese color. Después de tantos meses de preparativos, parece que el mundo entero está despertando despacio debajo de esta pequeña ventana junto a nosotros. Llevo un rato intentando pensar en algo inteligente que decir... pero sinceramente, me siento muy pequeño y un poco sobrepasado. Tú tampoco tienes que decir nada; pon la mano aquí y vamos a mirar juntos cómo cambia la luz durante un momento. Cuando volvamos a casa olvidaré la forma exacta de las nubes, pero creo que nunca olvidaré haber compartido este instante contigo.

Exact directing prompt; spoken text is unchanged
GENERAL: Clean, close-miked studio speech with one speaker. Natural, organic, spontaneous acting, not an advertisement or a formal announcement. Let the emotion evolve rather than staying fixed; keep small human hesitations and breaths. Breathless awe, a little vulnerability and quiet, shared tenderness.
SCRIPT:
(breathless wonder, almost a whisper) [6.4 seconds duration] Mira el borde de las nubes, justo ahí; sabía que el amanecer sería luminoso, pero no me esperaba ese color.
[0.3 seconds pause]
(expansive awe, gradually opening the voice) [6.5 seconds duration] Después de tantos meses de preparativos, parece que el mundo entero está despertando despacio debajo de esta pequeña ventana junto a nosotros.
[0.3 seconds pause]
(soft amazed exhale, 0.4 seconds)
(vulnerable honesty, a tiny searching pause) [6.8 seconds duration] Llevo un rato intentando pensar en algo inteligente que decir... pero sinceramente, me siento muy pequeño y un poco sobrepasado.
[0.3 seconds pause]
(soft affection and reassurance) [6.8 seconds duration] Tú tampoco tienes que decir nada; pon la mano aquí y vamos a mirar juntos cómo cambia la luz durante un momento.
[0.3 seconds pause]
(quiet inspiration, intimate rather than grandiose) [6.7 seconds duration] Cuando volvamos a casa olvidaré la forma exacta de las nubes, pero creo que nunca olvidaré haber compartido este instante contigo.

Fixed seed 20261627 · requested 435 frames · same separate reference

Published 1.5 Alpha · unmerged original

CE 6.25 · genuineness 4.43 · WER 0.0%
Fixed-target emotion MSE 1.897 (normalized 0.1186)

1.5 Alpha · merged, no additional SFT

CE 6.23 · genuineness 4.09 · WER 1.9%
Fixed-target emotion MSE 1.984 (normalized 0.1240)

1.5 Alpha + real-speech SFT · rank 1

CE 6.28 · genuineness 4.10 · WER 0.0%
Fixed-target emotion MSE 2.692 (normalized 0.1682)

1.5 Alpha + real-speech SFT · rank 2

CE 6.23 · genuineness 4.13 · WER 0.0%
Fixed-target emotion MSE 2.638 (normalized 0.1649)

1.5 Alpha + real-speech SFT · rank 4

CE 6.26 · genuineness 4.12 · WER 0.0%
Fixed-target emotion MSE 2.170 (normalized 0.1356)

1.5 Alpha + real-speech SFT · rank 8

CE 6.23 · genuineness 4.36 · WER 0.0%
Fixed-target emotion MSE 2.368 (normalized 0.1480)