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: Use manual-only configuration; no clear qualifying incremental adapter gain. Bucket scale 0.0, CFG 3.0.
| Held-out condition | n | Target z↑ | Raw head | CE↑ | Genuine↑ | Voice↑ | WER↓ | Valid | QA |
|---|---|---|---|---|---|---|---|---|---|
| standard | 50 | -0.353 | 0.170 | 6.237 | 4.358 | 0.579 | 0.013 | 1.000 | pass |
| manual | 50 | 1.922 | 2.030 | 5.950 | 4.089 | 0.533 | 0.030 | 0.940 | pass |
| adapter | 50 | 3.283 | 3.142 | 5.655 | 3.731 | 0.461 | 0.093 | 0.840 | fail |
| matched_no_adapter | 50 | 3.219 | 3.090 | 5.745 | 3.832 | 0.479 | 0.081 | 0.840 | fail |
Incremental adapter gain versus same-CFG control: 0.064 z;95% cluster interval [-0.1278124896315697, 0.2674999611696588]. Versus best manual-only: 1.361 z. Manual-only versus standard: 2.275 z.
| Scale | CFG | Target z | CE | Genuine | WER | Provisional QA |
|---|---|---|---|---|---|---|
| 0.0 | 1.0 | -0.134 | 6.141 | 4.529 | 0.000 | True |
| 0.5 | 1.0 | -0.203 | 6.115 | 4.416 | 0.008 | True |
| 0.75 | 1.0 | -0.153 | 6.112 | 4.428 | 0.003 | True |
| 1.0 | 1.0 | -0.173 | 6.115 | 4.389 | 0.000 | True |
| 0.0 | 1.25 | 0.077 | 6.127 | 4.461 | 0.003 | True |
| 0.5 | 1.25 | -0.068 | 6.123 | 4.472 | 0.030 | True |
| 0.75 | 1.25 | -0.070 | 6.094 | 4.419 | 0.008 | True |
| 1.0 | 1.25 | 0.246 | 6.056 | 4.392 | 0.014 | True |
| 0.0 | 1.5 | 0.097 | 6.083 | 4.398 | 0.008 | True |
| 0.5 | 1.5 | 0.323 | 6.123 | 4.504 | 0.009 | True |
| 0.75 | 1.5 | 0.114 | 6.100 | 4.460 | 0.011 | True |
| 1.0 | 1.5 | 0.306 | 6.052 | 4.365 | 0.008 | True |
| 0.0 | 2.0 | 0.864 | 6.092 | 4.461 | 0.034 | True |
| 0.5 | 2.0 | 1.041 | 6.056 | 4.358 | 0.016 | True |
| 0.75 | 2.0 | 0.784 | 6.030 | 4.367 | 0.049 | True |
| 1.0 | 2.0 | 1.196 | 5.876 | 4.246 | 0.025 | True |
| 0.0 | 2.5 | 1.934 | 6.018 | 4.248 | 0.014 | True |
| 0.5 | 2.5 | 1.585 | 5.839 | 4.187 | 0.014 | True |
| 0.75 | 2.5 | 1.647 | 5.942 | 4.167 | 0.022 | True |
| 1.0 | 2.5 | 1.805 | 5.914 | 4.113 | 0.079 | False |
| 0.0 | 3.0 | 2.470 | 5.854 | 4.100 | 0.022 | True |
| 0.5 | 3.0 | 2.136 | 5.732 | 4.125 | 0.029 | True |
| 0.75 | 3.0 | 2.268 | 5.766 | 4.076 | 0.013 | True |
| 1.0 | 3.0 | 1.956 | 5.804 | 4.073 | 0.080 | False |
| 0.0 | 4.0 | 2.837 | 5.818 | 4.034 | 0.115 | False |
| 0.5 | 4.0 | 2.676 | 5.677 | 3.947 | 0.105 | False |
| 0.75 | 4.0 | 2.917 | 5.634 | 3.875 | 0.103 | False |
| 1.0 | 4.0 | 2.911 | 5.485 | 3.846 | 0.035 | True |
| 0.0 | 5.0 | 3.070 | 5.711 | 3.921 | 0.080 | False |
| 0.5 | 5.0 | 3.137 | 5.616 | 3.840 | 0.051 | True |
| 0.75 | 5.0 | 3.198 | 5.644 | 3.830 | 0.102 | False |
| 1.0 | 5.0 | 3.434 | 5.336 | 3.646 | 0.148 | False |
direct_manual_recipe
{
"id": "manual-e66580c21865",
"general": "Emphasize vocal constriction and sharp intakes of breath to convey immediate panic. Pitch should fluctuate erratically with rising urgency.",
"inline": [
"Start with a tight, constricted vocal delivery suggesting immediate pressure building internally.",
"Maintain rapid, shallow breathing patterns throughout the utterance, making the words feel rushed and breathless.",
"Conclude with a sudden, sharp vocal break or catch in the breath to signal complete emotional overflow."
],
"cfg": 3,
"recipe": "P06",
"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 Distress. 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: {"es": 47, "nl": 87, "en": 59, "fr": 58, "de": 63}. 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 | 0.362 ± 0.055 | 6.188 ± 0.119 | 4.366 ± 0.182 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 20 | 0.358 ± 0.035 | 6.155 ± 0.156 | 4.331 ± 0.238 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 20 | 0.368 ± 0.076 | 6.203 ± 0.153 | 4.342 ± 0.224 | 0.26% ± 1.18 |
| Rank32 × 1.0 | 20 | 0.375 ± 0.066 | 6.169 ± 0.172 | 4.328 ± 0.214 | 4.64% ± 18.44 |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.352 ± 0.024 | 6.303 ± 0.063 | 4.266 ± 0.152 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.355 ± 0.023 | 6.221 ± 0.144 | 4.172 ± 0.399 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.334 ± 0.030 | 6.277 ± 0.066 | 4.148 ± 0.162 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.358 ± 0.022 | 6.269 ± 0.050 | 4.296 ± 0.096 | 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.387 · CE 6.28 · Genuine 4.14 · WER 0.0% | Target 0.386 · CE 6.28 · Genuine 4.27 · WER 0.0% | Target 0.329 · CE 6.22 · Genuine 3.97 · WER 0.0% | Target 0.374 · CE 6.24 · Genuine 4.19 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.345 · CE 6.26 · Genuine 4.29 · WER 0.0% | Target 0.357 · CE 6.05 · Genuine 3.59 · WER 0.0% | Target 0.377 · CE 6.23 · Genuine 4.17 · WER 0.0% | Target 0.377 · CE 6.22 · Genuine 4.26 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 4.47 · WER 0.0% | Target 0.347 · CE 6.38 · Genuine 4.51 · WER 0.0% | Target 0.315 · CE 6.36 · Genuine 4.36 · WER 0.0% | Target 0.351 · CE 6.28 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.338 · CE 6.28 · Genuine 4.16 · WER 0.0% | Target 0.332 · CE 6.17 · Genuine 4.32 · WER 0.0% | Target 0.312 · CE 6.29 · Genuine 4.10 · WER 0.0% | Target 0.329 · CE 6.33 · Genuine 4.42 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.337 ± 0.018 | 6.299 ± 0.083 | 4.449 ± 0.084 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.342 ± 0.019 | 6.360 ± 0.127 | 4.510 ± 0.080 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.344 ± 0.045 | 6.384 ± 0.085 | 4.432 ± 0.197 | 1.32% ± 2.63 |
| Rank32 × 1.0 | 4 | 0.356 ± 0.048 | 6.344 ± 0.164 | 4.512 ± 0.178 | 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.357 · CE 6.23 · Genuine 4.47 · WER 0.0% | Target 0.363 · CE 6.28 · Genuine 4.44 · WER 0.0% | Target 0.376 · CE 6.36 · Genuine 4.49 · WER 5.3% | Target 0.412 · CE 6.12 · Genuine 4.51 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.331 · CE 6.27 · Genuine 4.41 · WER 0.0% | Target 0.349 · CE 6.23 · Genuine 4.62 · WER 0.0% | Target 0.389 · CE 6.28 · Genuine 4.68 · WER 0.0% | Target 0.378 · CE 6.36 · Genuine 4.76 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen. | Target 0.342 · CE 6.42 · Genuine 4.56 · WER 0.0% | Target 0.320 · CE 6.51 · Genuine 4.46 · WER 0.0% | Target 0.310 · CE 6.48 · Genuine 4.29 · WER 0.0% | Target 0.324 · CE 6.40 · Genuine 4.44 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen. | Target 0.315 · CE 6.28 · Genuine 4.36 · WER 0.0% | Target 0.334 · CE 6.42 · Genuine 4.52 · WER 0.0% | Target 0.302 · CE 6.42 · Genuine 4.26 · WER 0.0% | Target 0.309 · CE 6.51 · Genuine 4.34 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.345 ± 0.017 | 6.144 ± 0.072 | 4.413 ± 0.218 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.349 ± 0.016 | 6.017 ± 0.073 | 4.401 ± 0.148 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.361 ± 0.011 | 6.168 ± 0.063 | 4.459 ± 0.131 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.356 ± 0.012 | 6.122 ± 0.068 | 4.460 ± 0.262 | 2.63% ± 5.26 |
| 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.352 · CE 6.08 · Genuine 4.32 · WER 0.0% | Target 0.364 · CE 5.97 · Genuine 4.24 · WER 0.0% | Target 0.358 · CE 6.13 · Genuine 4.38 · WER 0.0% | Target 0.368 · CE 6.02 · Genuine 4.12 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.344 · CE 6.14 · Genuine 4.16 · WER 0.0% | Target 0.363 · CE 5.94 · Genuine 4.31 · WER 0.0% | Target 0.376 · CE 6.14 · Genuine 4.34 · WER 0.0% | Target 0.361 · CE 6.13 · Genuine 4.43 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.321 · CE 6.25 · Genuine 4.53 · WER 0.0% | Target 0.338 · CE 6.06 · Genuine 4.51 · WER 0.0% | Target 0.360 · CE 6.26 · Genuine 4.63 · WER 0.0% | Target 0.341 · CE 6.17 · Genuine 4.55 · WER 10.5% |
| 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.361 · CE 6.11 · Genuine 4.65 · WER 0.0% | Target 0.332 · CE 6.09 · Genuine 4.55 · WER 0.0% | Target 0.349 · CE 6.14 · Genuine 4.48 · WER 0.0% | Target 0.352 · CE 6.16 · Genuine 4.75 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.426 ± 0.102 | 6.098 ± 0.069 | 4.269 ± 0.089 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.411 ± 0.039 | 6.116 ± 0.082 | 4.256 ± 0.141 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.472 ± 0.119 | 6.040 ± 0.195 | 4.186 ± 0.150 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.473 ± 0.086 | 6.183 ± 0.036 | 4.275 ± 0.147 | 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.569 · CE 6.17 · Genuine 4.34 · WER 0.0% | Target 0.466 · CE 6.07 · Genuine 4.36 · WER 0.0% | Target 0.442 · CE 6.11 · Genuine 4.25 · WER 0.0% | Target 0.568 · CE 6.14 · Genuine 4.32 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.428 · CE 6.11 · Genuine 4.17 · WER 0.0% | Target 0.402 · CE 6.06 · Genuine 4.17 · WER 0.0% | Target 0.394 · CE 5.75 · Genuine 4.10 · WER 0.0% | Target 0.508 · CE 6.21 · Genuine 4.21 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.341 · CE 6.10 · Genuine 4.22 · WER 0.0% | Target 0.373 · CE 6.09 · Genuine 4.11 · WER 0.0% | Target 0.648 · CE 6.18 · Genuine 4.03 · WER 0.0% | Target 0.448 · CE 6.22 · Genuine 4.46 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. 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 Distress. 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 Distress. 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 0.367 · CE 6.01 · Genuine 4.34 · WER 0.0% | Target 0.402 · CE 6.24 · Genuine 4.39 · WER 0.0% | Target 0.404 · CE 6.12 · Genuine 4.36 · WER 0.0% | Target 0.367 · CE 6.17 · Genuine 4.12 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.351 ± 0.027 | 6.096 ± 0.102 | 4.435 ± 0.283 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.335 ± 0.009 | 6.062 ± 0.066 | 4.313 ± 0.248 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.332 ± 0.018 | 6.146 ± 0.046 | 4.487 ± 0.269 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.333 ± 0.020 | 5.926 ± 0.128 | 4.099 ± 0.120 | 20.59% ± 41.18 |
| 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta. (Intense Distress. 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo. | Target 0.390 · CE 6.04 · Genuine 4.44 · WER 0.0% | Target 0.327 · CE 5.98 · Genuine 4.03 · WER 0.0% | Target 0.327 · CE 6.15 · Genuine 4.33 · WER 0.0% | Target 0.336 · CE 5.98 · Genuine 4.20 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta. (Intense Distress. 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo. | Target 0.345 · CE 6.00 · Genuine 4.04 · WER 0.0% | Target 0.342 · CE 6.04 · Genuine 4.27 · WER 0.0% | Target 0.331 · CE 6.15 · Genuine 4.24 · WER 0.0% | Target 0.339 · CE 5.90 · Genuine 4.01 · 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita. (Intense Distress. 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 Distress. 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 0.331 · CE 6.12 · Genuine 4.55 · WER 0.0% | Target 0.327 · CE 6.09 · Genuine 4.63 · WER 0.0% | Target 0.313 · CE 6.20 · Genuine 4.53 · WER 0.0% | Target 0.352 · CE 5.76 · Genuine 3.98 · WER 82.4% |
| 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 Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Distress. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita. (Intense Distress. 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 Distress. 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 0.337 · CE 6.23 · Genuine 4.71 · WER 0.0% | Target 0.344 · CE 6.13 · Genuine 4.32 · WER 0.0% | Target 0.355 · CE 6.09 · Genuine 4.85 · WER 0.0% | Target 0.304 · CE 6.06 · Genuine 4.21 · WER 0.0% |