The merged parent used in these specialist comparisons is now released as Humaneness Voice Large 1.6 Alpha: 1.5 + final RealSpeech rank 4, merged once. Apply only the selected new bucket adapter. Do not reload either ancestor.
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.
The original grid used generic GENERAL/SCRIPT directions at CFG1. It did not systematically use target-specific manual winners. The new study freezes the actual rank-4 manual wording, preserves every literal word, seed, reference, adapter and frame budget, then independently varies CFG and adapter strength. The previous recordings and conclusions remain below.
Ranking:100% requested EARS emotion/style head, standardized using frozen training statistics; low-pole targets reverse the sign. Formal/casual contrasts average the desired head and negative opposite head. CE and genuineness are diagnostics. Higher target z means stronger desired delivery, not a human rating. More CFG is not automatically better.
Selection:20 original-text outcomes per configuration select wording, CFG and scale. Fifty additional unseen-text reference pairs test the frozen choices. A recommended adapter must improve against both the same-CFG no-adapter control and the best manual-only setting. Quality gates: per-take WER≤15%, mean WER≤10%, valid/terminated fraction≥90%, paired identity change≥−0.05 over50pairs. Failed outcomes stay visible. Language-cluster bootstrap intervals use only five fixed reference voices; findings across72targets remain exploratory.
Five languages: EN/DE/FR/NL/ES; original seeds1701/1702; reference187frames/14.96seconds. CFG1/1.25/1.5/2/2.5/3/4/5 × bucket strength0/0.5/0.75/1. CFG4/5 are stress conditions, not presumed improvements. No-reference use of these bucket recommendations has not been validated.
Taxonomy:S_MONO means Monologue, S_TECH means Teacher/Didactic, VOLT means Volatility. The two corrected style recipes and17dimension/contrast recipes are new searches, not historical manual winners.53other targets reuse directly applicable manual recipes.
Best tested agent route: Held-out target improves against matched CFG and best manual-only configuration; QA passes. Bucket scale 0.5, CFG 5.0.
| Held-out condition | n | Target z↑ | Raw head | CE↑ | Genuine↑ | Voice↑ | WER↓ | Valid | QA |
|---|---|---|---|---|---|---|---|---|---|
| standard | 50 | -0.334 | 0.141 | 6.247 | 4.377 | 0.594 | 0.018 | 0.980 | pass |
| manual | 50 | 0.112 | 0.570 | 5.809 | 4.640 | 0.601 | 0.026 | 0.960 | pass |
| adapter | 50 | 0.159 | 0.615 | 5.848 | 4.647 | 0.610 | 0.023 | 1.000 | pass |
| matched_no_adapter | 50 | 0.112 | 0.570 | 5.809 | 4.640 | 0.601 | 0.026 | 0.960 | pass |
Incremental adapter gain versus same-CFG control: 0.047 z;95% cluster interval [0.013994134857910307, 0.10048585041987572]. Versus best manual-only: 0.047 z. Manual-only versus standard: 0.446 z.
| Scale | CFG | Target z | CE | Genuine | WER | Provisional QA |
|---|---|---|---|---|---|---|
| 0.0 | 1.0 | -0.313 | 6.183 | 4.499 | 0.000 | True |
| 0.5 | 1.0 | -0.301 | 6.097 | 4.450 | 0.006 | True |
| 0.75 | 1.0 | -0.291 | 6.130 | 4.439 | 0.003 | True |
| 1.0 | 1.0 | -0.310 | 6.165 | 4.510 | 0.008 | True |
| 0.0 | 1.25 | -0.267 | 6.104 | 4.531 | 0.008 | True |
| 0.5 | 1.25 | -0.236 | 6.125 | 4.498 | 0.000 | True |
| 0.75 | 1.25 | -0.238 | 6.117 | 4.459 | 0.000 | True |
| 1.0 | 1.25 | -0.282 | 6.117 | 4.502 | 0.029 | True |
| 0.0 | 1.5 | -0.213 | 6.109 | 4.499 | 0.007 | True |
| 0.5 | 1.5 | -0.263 | 6.105 | 4.554 | 0.006 | True |
| 0.75 | 1.5 | -0.239 | 6.094 | 4.521 | 0.024 | True |
| 1.0 | 1.5 | -0.217 | 6.099 | 4.529 | 0.003 | True |
| 0.0 | 2.0 | -0.208 | 6.005 | 4.523 | 0.000 | True |
| 0.5 | 2.0 | -0.160 | 6.062 | 4.555 | 0.017 | True |
| 0.75 | 2.0 | -0.183 | 6.040 | 4.601 | 0.006 | True |
| 1.0 | 2.0 | -0.190 | 6.015 | 4.553 | 0.003 | True |
| 0.0 | 2.5 | -0.118 | 6.032 | 4.593 | 0.003 | True |
| 0.5 | 2.5 | -0.044 | 5.968 | 4.593 | 0.003 | True |
| 0.75 | 2.5 | -0.081 | 5.961 | 4.654 | 0.006 | True |
| 1.0 | 2.5 | -0.119 | 5.971 | 4.586 | 0.009 | True |
| 0.0 | 3.0 | -0.062 | 5.928 | 4.596 | 0.006 | True |
| 0.5 | 3.0 | -0.029 | 5.969 | 4.620 | 0.003 | True |
| 0.75 | 3.0 | -0.004 | 5.945 | 4.584 | 0.005 | True |
| 1.0 | 3.0 | -0.006 | 5.929 | 4.599 | 0.003 | True |
| 0.0 | 4.0 | 0.060 | 5.841 | 4.623 | 0.015 | True |
| 0.5 | 4.0 | 0.093 | 5.815 | 4.612 | 0.006 | True |
| 0.75 | 4.0 | 0.033 | 5.806 | 4.614 | 0.008 | True |
| 1.0 | 4.0 | 0.037 | 5.804 | 4.613 | 0.000 | True |
| 0.0 | 5.0 | 0.101 | 5.706 | 4.556 | 0.008 | True |
| 0.5 | 5.0 | 0.119 | 5.661 | 4.596 | 0.000 | True |
| 0.75 | 5.0 | 0.113 | 5.690 | 4.587 | 0.005 | True |
| 1.0 | 5.0 | 0.071 | 5.719 | 4.624 | 0.006 | True |
direct_manual_recipe
{
"id": "manual-53d14c32a3a7",
"general": "Maintain a consistently flat, almost monotone pitch contour throughout the entire utterance, devoid of any natural inflections.",
"inline": [
"Initiate with a breathy, minimal vocal projection that suggests detachment.",
"Sustain a mid-range, unchanging vocal placement across the core of the statement.",
"Deliver the closing words with the same low energy, ensuring no rise or fall."
],
"cfg": 3,
"recipe": "P05",
"strategy": "local-Gemma-mutation",
"timing_syntax": "canonical",
"canonical_labels": true,
"neutral_event": "keep",
"parent": "cfg-3",
"origin": "exact_frozen_manual_recipe"
}Exact machine-readable comparison · Agent tool recommendations
Intense Emotional Numbness. Base: 1.5 Alpha + merged real-speech rank4; new adapter: rank32. Every row is a matched seed/text/reference test. Scale zero disables the new adapter; all other scales are applied absolutely, never compounded.
Adapter training data: {"nl": 6, "de": 6, "fr": 6, "en": 6, "es": 7}. All available accepted languages were pooled, globally shuffled, for three epochs. Peak LR 2e-05 at the end of the first-epoch warmup, then cosine decay. All original optimizer/RNG checkpoints remain saved.
Taxonomy: S_MONO is monologue, not monotone; VOLT is volatility, not volume. No claim is made that a classifier score alone proves the audible performance.
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 20 | 0.359 ± 0.025 | 6.201 ± 0.136 | 4.334 ± 0.213 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 20 | 0.361 ± 0.021 | 6.213 ± 0.131 | 4.307 ± 0.200 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 20 | 0.367 ± 0.027 | 6.185 ± 0.148 | 4.339 ± 0.263 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 20 | 0.360 ± 0.023 | 6.188 ± 0.126 | 4.295 ± 0.265 | 0.00% ± 0.00 |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.347 ± 0.016 | 6.315 ± 0.061 | 4.264 ± 0.346 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.350 ± 0.017 | 6.328 ± 0.139 | 4.145 ± 0.336 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.339 ± 0.017 | 6.281 ± 0.160 | 4.095 ± 0.343 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.342 ± 0.014 | 6.297 ± 0.073 | 4.030 ± 0.284 | 0.00% ± 0.00 |
| Prompt / seed | Merged base · no new LoRA | Rank32 × 0.5 | Rank32 × 0.75 | Rank32 × 1.0 |
|---|---|---|---|---|
| Seed 1701 The door is still open. I put your coat on the chair. Tell me when you are ready. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready. | Target 0.370 · CE 6.28 · Genuine 4.28 · WER 0.0% | Target 0.372 · CE 6.19 · Genuine 3.93 · WER 0.0% | Target 0.353 · CE 6.24 · Genuine 3.69 · WER 0.0% | Target 0.354 · CE 6.19 · Genuine 3.67 · WER 0.0% |
| Seed 1702 The door is still open. I put your coat on the chair. Tell me when you are ready. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready. | Target 0.339 · CE 6.28 · Genuine 3.77 · WER 0.0% | Target 0.356 · CE 6.25 · Genuine 3.79 · WER 0.0% | Target 0.354 · CE 6.08 · Genuine 3.93 · WER 0.0% | Target 0.349 · CE 6.32 · Genuine 3.96 · WER 0.0% |
| Seed 1701 I found the little box. It was beside the old window. We can look inside together. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together. | Target 0.337 · CE 6.40 · Genuine 4.44 · WER 0.0% | Target 0.337 · CE 6.36 · Genuine 4.38 · WER 0.0% | Target 0.320 · CE 6.44 · Genuine 4.36 · WER 0.0% | Target 0.323 · CE 6.35 · Genuine 4.31 · WER 0.0% |
| Seed 1702 I found the little box. It was beside the old window. We can look inside together. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together. | Target 0.340 · CE 6.30 · Genuine 4.56 · WER 0.0% | Target 0.337 · CE 6.51 · Genuine 4.48 · WER 0.0% | Target 0.329 · CE 6.36 · Genuine 4.40 · WER 0.0% | Target 0.340 · CE 6.33 · Genuine 4.19 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.358 ± 0.018 | 6.332 ± 0.137 | 4.388 ± 0.182 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.356 ± 0.025 | 6.307 ± 0.085 | 4.272 ± 0.154 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.354 ± 0.021 | 6.293 ± 0.103 | 4.267 ± 0.129 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.356 ± 0.017 | 6.259 ± 0.104 | 4.399 ± 0.149 | 0.00% ± 0.00 |
| Prompt / seed | Merged base · no new LoRA | Rank32 × 0.5 | Rank32 × 0.75 | Rank32 × 1.0 |
|---|---|---|---|---|
| Seed 1701 Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist. | Target 0.372 · CE 6.14 · Genuine 4.62 · WER 0.0% | Target 0.371 · CE 6.23 · Genuine 4.26 · WER 0.0% | Target 0.368 · CE 6.21 · Genuine 4.38 · WER 0.0% | Target 0.374 · CE 6.21 · Genuine 4.61 · WER 0.0% |
| Seed 1702 Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist. | Target 0.373 · CE 6.32 · Genuine 4.33 · WER 0.0% | Target 0.382 · CE 6.24 · Genuine 4.47 · WER 0.0% | Target 0.374 · CE 6.20 · Genuine 4.33 · WER 0.0% | Target 0.363 · CE 6.18 · Genuine 4.38 · WER 0.0% |
| Seed 1701 Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen. | Target 0.337 · CE 6.43 · Genuine 4.19 · WER 0.0% | Target 0.334 · CE 6.38 · Genuine 4.09 · WER 0.0% | Target 0.329 · CE 6.40 · Genuine 4.09 · WER 0.0% | Target 0.349 · CE 6.41 · Genuine 4.35 · WER 0.0% |
| Seed 1702 Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen. | Target 0.349 · CE 6.44 · Genuine 4.42 · WER 0.0% | Target 0.335 · CE 6.38 · Genuine 4.27 · WER 0.0% | Target 0.344 · CE 6.36 · Genuine 4.27 · WER 0.0% | Target 0.337 · CE 6.24 · Genuine 4.26 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.338 ± 0.005 | 6.177 ± 0.081 | 4.501 ± 0.135 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.347 ± 0.011 | 6.128 ± 0.071 | 4.441 ± 0.035 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.383 ± 0.018 | 6.185 ± 0.065 | 4.636 ± 0.100 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.349 ± 0.018 | 6.225 ± 0.086 | 4.576 ± 0.223 | 0.00% ± 0.00 |
| Prompt / seed | Merged base · no new LoRA | Rank32 × 0.5 | Rank32 × 0.75 | Rank32 × 1.0 |
|---|---|---|---|---|
| Seed 1701 La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt. | Target 0.342 · CE 6.17 · Genuine 4.62 · WER 0.0% | Target 0.363 · CE 6.21 · Genuine 4.48 · WER 0.0% | Target 0.409 · CE 6.21 · Genuine 4.71 · WER 0.0% | Target 0.352 · CE 6.10 · Genuine 4.28 · WER 0.0% |
| Seed 1702 La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt. | Target 0.342 · CE 6.29 · Genuine 4.48 · WER 0.0% | Target 0.345 · CE 6.16 · Genuine 4.41 · WER 0.0% | Target 0.368 · CE 6.09 · Genuine 4.51 · WER 0.0% | Target 0.364 · CE 6.28 · Genuine 4.78 · WER 0.0% |
| Seed 1701 J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient. | Target 0.332 · CE 6.11 · Genuine 4.58 · WER 0.0% | Target 0.344 · CE 6.11 · Genuine 4.42 · WER 0.0% | Target 0.374 · CE 6.25 · Genuine 4.61 · WER 0.0% | Target 0.324 · CE 6.28 · Genuine 4.53 · WER 0.0% |
| Seed 1702 J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient. | Target 0.334 · CE 6.13 · Genuine 4.32 · WER 0.0% | Target 0.337 · CE 6.04 · Genuine 4.46 · WER 0.0% | Target 0.381 · CE 6.19 · Genuine 4.72 · WER 0.0% | Target 0.358 · CE 6.23 · Genuine 4.71 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.393 ± 0.030 | 6.048 ± 0.079 | 4.194 ± 0.113 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.383 ± 0.014 | 6.162 ± 0.082 | 4.311 ± 0.086 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.389 ± 0.014 | 6.102 ± 0.060 | 4.291 ± 0.117 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.372 ± 0.016 | 6.096 ± 0.106 | 4.150 ± 0.090 | 0.00% ± 0.00 |
| Prompt / seed | Merged base · no new LoRA | Rank32 × 0.5 | Rank32 × 0.75 | Rank32 × 1.0 |
|---|---|---|---|---|
| Seed 1701 De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent. | Target 0.386 · CE 6.03 · Genuine 4.13 · WER 0.0% | Target 0.382 · CE 6.13 · Genuine 4.33 · WER 0.0% | Target 0.374 · CE 6.15 · Genuine 4.32 · WER 0.0% | Target 0.349 · CE 6.19 · Genuine 4.21 · WER 0.0% |
| Seed 1702 De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent. | Target 0.436 · CE 5.96 · Genuine 4.11 · WER 0.0% | Target 0.368 · CE 6.13 · Genuine 4.19 · WER 0.0% | Target 0.406 · CE 6.01 · Genuine 4.32 · WER 0.0% | Target 0.377 · CE 6.09 · Genuine 4.13 · WER 0.0% |
| Seed 1701 Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken. | Target 0.379 · CE 6.15 · Genuine 4.36 · WER 0.0% | Target 0.401 · CE 6.11 · Genuine 4.32 · WER 0.0% | Target 0.382 · CE 6.11 · Genuine 4.12 · WER 0.0% | Target 0.383 · CE 5.95 · Genuine 4.03 · WER 0.0% |
| Seed 1702 Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken. | Target 0.369 · CE 6.05 · Genuine 4.18 · WER 0.0% | Target 0.380 · CE 6.28 · Genuine 4.40 · WER 0.0% | Target 0.393 · CE 6.13 · Genuine 4.40 · WER 0.0% | Target 0.381 · CE 6.16 · Genuine 4.23 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.358 ± 0.016 | 6.132 ± 0.057 | 4.321 ± 0.171 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.367 ± 0.021 | 6.142 ± 0.143 | 4.366 ± 0.208 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.370 ± 0.035 | 6.065 ± 0.196 | 4.408 ± 0.258 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.382 ± 0.031 | 6.064 ± 0.101 | 4.319 ± 0.201 | 0.00% ± 0.00 |
| Prompt / seed | Merged base · no new LoRA | Rank32 × 0.5 | Rank32 × 0.75 | Rank32 × 1.0 |
|---|---|---|---|---|
| Seed 1701 La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo. | Target 0.347 · CE 6.05 · Genuine 4.18 · WER 0.0% | Target 0.360 · CE 6.25 · Genuine 4.36 · WER 0.0% | Target 0.418 · CE 6.18 · Genuine 4.09 · WER 0.0% | Target 0.425 · CE 6.10 · Genuine 4.14 · WER 0.0% |
| Seed 1702 La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo. | Target 0.374 · CE 6.15 · Genuine 4.19 · WER 0.0% | Target 0.396 · CE 5.97 · Genuine 4.09 · WER 0.0% | Target 0.375 · CE 5.77 · Genuine 4.34 · WER 0.0% | Target 0.380 · CE 5.97 · Genuine 4.16 · WER 0.0% |
| Seed 1701 He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro. | Target 0.342 · CE 6.17 · Genuine 4.36 · WER 0.0% | Target 0.366 · CE 6.07 · Genuine 4.42 · WER 0.0% | Target 0.342 · CE 6.13 · Genuine 4.70 · WER 0.0% | Target 0.349 · CE 6.19 · Genuine 4.43 · WER 0.0% |
| Seed 1702 He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro. Exact GENERAL / SCRIPT promptGENERAL: One adult speaker in a clean recording. Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja. (Intense Emotional Numbness. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro. | Target 0.369 · CE 6.17 · Genuine 4.55 · WER 0.0% | Target 0.347 · CE 6.27 · Genuine 4.60 · WER 0.0% | Target 0.346 · CE 6.17 · Genuine 4.51 · WER 0.0% | Target 0.374 · CE 5.99 · Genuine 4.55 · WER 0.0% |