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 2.5.
| Held-out condition | n | Target z↑ | Raw head | CE↑ | Genuine↑ | Voice↑ | WER↓ | Valid | QA |
|---|---|---|---|---|---|---|---|---|---|
| standard | 50 | -0.275 | 0.050 | 6.275 | 4.376 | 0.606 | 0.010 | 1.000 | pass |
| manual | 50 | -0.016 | 0.190 | 6.060 | 4.504 | 0.575 | 0.008 | 1.000 | pass |
| adapter | 50 | -0.043 | 0.175 | 5.869 | 4.407 | 0.531 | 0.013 | 1.000 | fail |
| matched_no_adapter | 50 | 0.030 | 0.215 | 5.879 | 4.427 | 0.537 | 0.036 | 0.980 | fail |
Incremental adapter gain versus same-CFG control: -0.073 z;95% cluster interval [-0.13878663475989475, -0.007316897094878844]. Versus best manual-only: -0.027 z. Manual-only versus standard: 0.259 z.
| Scale | CFG | Target z | CE | Genuine | WER | Provisional QA |
|---|---|---|---|---|---|---|
| 0.0 | 1.0 | -0.275 | 6.189 | 4.479 | 0.003 | True |
| 0.5 | 1.0 | -0.232 | 6.199 | 4.451 | 0.000 | True |
| 0.75 | 1.0 | -0.250 | 6.166 | 4.351 | 0.000 | True |
| 1.0 | 1.0 | -0.270 | 6.225 | 4.439 | 0.008 | False |
| 0.0 | 1.25 | -0.245 | 6.167 | 4.471 | 0.000 | True |
| 0.5 | 1.25 | -0.222 | 6.201 | 4.469 | 0.000 | True |
| 0.75 | 1.25 | -0.225 | 6.155 | 4.479 | 0.003 | True |
| 1.0 | 1.25 | -0.270 | 6.164 | 4.428 | 0.009 | True |
| 0.0 | 1.5 | -0.206 | 6.148 | 4.470 | 0.000 | True |
| 0.5 | 1.5 | -0.244 | 6.189 | 4.500 | 0.000 | True |
| 0.75 | 1.5 | -0.191 | 6.172 | 4.496 | 0.003 | True |
| 1.0 | 1.5 | -0.186 | 6.156 | 4.494 | 0.026 | True |
| 0.0 | 2.0 | -0.019 | 6.055 | 4.472 | 0.006 | False |
| 0.5 | 2.0 | -0.185 | 6.059 | 4.488 | 0.005 | True |
| 0.75 | 2.0 | -0.089 | 6.078 | 4.468 | 0.000 | True |
| 1.0 | 2.0 | -0.112 | 6.063 | 4.523 | 0.005 | False |
| 0.0 | 2.5 | 0.041 | 6.021 | 4.510 | 0.003 | True |
| 0.5 | 2.5 | 0.054 | 6.042 | 4.501 | 0.006 | False |
| 0.75 | 2.5 | -0.069 | 5.967 | 4.434 | 0.008 | True |
| 1.0 | 2.5 | 0.013 | 5.989 | 4.494 | 0.006 | True |
| 0.0 | 3.0 | 0.107 | 5.894 | 4.473 | 0.003 | False |
| 0.5 | 3.0 | 0.098 | 5.908 | 4.473 | 0.003 | False |
| 0.75 | 3.0 | 0.060 | 5.835 | 4.429 | 0.000 | False |
| 1.0 | 3.0 | 0.097 | 5.800 | 4.410 | 0.015 | True |
| 0.0 | 4.0 | 0.131 | 5.550 | 4.343 | 0.011 | False |
| 0.5 | 4.0 | 0.120 | 5.687 | 4.390 | 0.003 | False |
| 0.75 | 4.0 | 0.091 | 5.598 | 4.399 | 0.005 | False |
| 1.0 | 4.0 | 0.312 | 5.664 | 4.399 | 0.000 | False |
| 0.0 | 5.0 | 0.095 | 5.518 | 4.354 | 0.023 | False |
| 0.5 | 5.0 | 0.249 | 5.541 | 4.344 | 0.024 | False |
| 0.75 | 5.0 | 0.206 | 5.510 | 4.344 | 0.012 | False |
| 1.0 | 5.0 | 0.161 | 5.528 | 4.361 | 0.003 | False |
direct_manual_recipe
{
"id": "manual-e0a2c5b12b00",
"general": "Manipulate vocal projection subtly; the emotion should be felt more than loudly stated, a warm intimacy.",
"inline": [
"Begin at a hushed, intimate volume, as if sharing a secret only with the listener.",
"Allow the volume to swell ever so slightly mid-phrase, a momentary rush of feeling.",
"Bring the volume back down to a near whisper for the final words, sealing the private moment."
],
"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 Infatuation. 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: {"en": 23, "de": 5, "es": 3, "fr": 1}. 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 | 0.376 ± 0.027 | 6.188 ± 0.102 | 4.387 ± 0.190 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 20 | 0.381 ± 0.035 | 6.163 ± 0.126 | 4.305 ± 0.222 | 0.26% ± 1.18 |
| Rank32 × 0.75 | 20 | 0.380 ± 0.031 | 6.183 ± 0.147 | 4.333 ± 0.251 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 20 | 0.377 ± 0.034 | 6.167 ± 0.127 | 4.364 ± 0.247 | 0.00% ± 0.00 |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.409 ± 0.020 | 6.267 ± 0.087 | 4.299 ± 0.148 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.422 ± 0.037 | 6.285 ± 0.064 | 4.310 ± 0.220 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.426 ± 0.029 | 6.301 ± 0.117 | 4.307 ± 0.250 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.403 ± 0.044 | 6.307 ± 0.064 | 4.300 ± 0.421 | 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.414 · CE 6.23 · Genuine 4.32 · WER 0.0% | Target 0.458 · CE 6.25 · Genuine 4.12 · WER 0.0% | Target 0.416 · CE 6.19 · Genuine 4.22 · WER 0.0% | Target 0.358 · CE 6.25 · Genuine 3.96 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.379 · CE 6.17 · Genuine 4.09 · WER 0.0% | Target 0.371 · CE 6.25 · Genuine 4.13 · WER 0.0% | Target 0.389 · CE 6.21 · Genuine 3.99 · WER 0.0% | Target 0.375 · CE 6.25 · Genuine 3.92 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.421 · CE 6.36 · Genuine 4.44 · WER 0.0% | Target 0.435 · CE 6.38 · Genuine 4.56 · WER 0.0% | Target 0.448 · CE 6.43 · Genuine 4.52 · WER 0.0% | Target 0.453 · CE 6.38 · Genuine 4.67 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.420 · CE 6.31 · Genuine 4.34 · WER 0.0% | Target 0.426 · CE 6.26 · Genuine 4.43 · WER 0.0% | Target 0.449 · CE 6.37 · Genuine 4.49 · WER 0.0% | Target 0.427 · CE 6.34 · Genuine 4.66 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.348 ± 0.026 | 6.273 ± 0.071 | 4.327 ± 0.308 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.347 ± 0.026 | 6.289 ± 0.111 | 4.314 ± 0.146 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.366 ± 0.024 | 6.289 ± 0.203 | 4.280 ± 0.197 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.356 ± 0.040 | 6.273 ± 0.033 | 4.457 ± 0.045 | 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.318 · CE 6.25 · Genuine 4.51 · WER 0.0% | Target 0.319 · CE 6.18 · Genuine 4.40 · WER 0.0% | Target 0.335 · CE 6.33 · Genuine 4.51 · WER 0.0% | Target 0.344 · CE 6.31 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.334 · CE 6.18 · Genuine 4.22 · WER 0.0% | Target 0.342 · CE 6.21 · Genuine 4.47 · WER 0.0% | Target 0.370 · CE 6.02 · Genuine 4.05 · WER 0.0% | Target 0.307 · CE 6.25 · Genuine 4.48 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen. | Target 0.364 · CE 6.33 · Genuine 3.94 · WER 0.0% | Target 0.381 · CE 6.40 · Genuine 4.23 · WER 0.0% | Target 0.393 · CE 6.52 · Genuine 4.36 · WER 0.0% | Target 0.372 · CE 6.25 · Genuine 4.42 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen. | Target 0.374 · CE 6.32 · Genuine 4.63 · WER 0.0% | Target 0.347 · CE 6.37 · Genuine 4.16 · WER 0.0% | Target 0.364 · CE 6.28 · Genuine 4.21 · WER 0.0% | Target 0.401 · CE 6.29 · Genuine 4.42 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.384 ± 0.030 | 6.152 ± 0.085 | 4.589 ± 0.125 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.391 ± 0.025 | 6.136 ± 0.032 | 4.379 ± 0.134 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.380 ± 0.030 | 6.114 ± 0.109 | 4.507 ± 0.088 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.385 ± 0.041 | 6.110 ± 0.035 | 4.341 ± 0.315 | 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.355 · CE 6.17 · Genuine 4.49 · WER 0.0% | Target 0.364 · CE 6.13 · Genuine 4.32 · WER 0.0% | Target 0.357 · CE 6.17 · Genuine 4.48 · WER 0.0% | Target 0.339 · CE 6.13 · Genuine 4.19 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.362 · CE 6.14 · Genuine 4.63 · WER 0.0% | Target 0.376 · CE 6.11 · Genuine 4.39 · WER 0.0% | Target 0.352 · CE 6.06 · Genuine 4.39 · WER 0.0% | Target 0.362 · CE 6.07 · Genuine 4.60 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.410 · CE 6.05 · Genuine 4.48 · WER 0.0% | Target 0.407 · CE 6.12 · Genuine 4.56 · WER 0.0% | Target 0.410 · CE 5.99 · Genuine 4.58 · WER 0.0% | Target 0.427 · CE 6.15 · Genuine 4.61 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.409 · CE 6.25 · Genuine 4.75 · WER 0.0% | Target 0.416 · CE 6.18 · Genuine 4.24 · WER 0.0% | Target 0.402 · CE 6.24 · Genuine 4.57 · WER 0.0% | Target 0.412 · CE 6.09 · Genuine 3.97 · WER 0.0% |
No accepted bucket-training example in this language: transfer evaluation only.
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.371 ± 0.008 | 6.110 ± 0.091 | 4.349 ± 0.068 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.380 ± 0.016 | 6.090 ± 0.068 | 4.222 ± 0.200 | 1.32% ± 2.63 |
| Rank32 × 0.75 | 4 | 0.369 ± 0.009 | 6.130 ± 0.043 | 4.245 ± 0.116 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.374 ± 0.008 | 6.100 ± 0.084 | 4.352 ± 0.136 | 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.383 · CE 6.11 · Genuine 4.27 · WER 0.0% | Target 0.402 · CE 6.13 · Genuine 4.13 · WER 0.0% | Target 0.364 · CE 6.07 · Genuine 4.34 · WER 0.0% | Target 0.363 · CE 6.01 · Genuine 4.36 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.368 · CE 6.11 · Genuine 4.39 · WER 0.0% | Target 0.373 · CE 6.00 · Genuine 4.30 · WER 5.3% | Target 0.378 · CE 6.17 · Genuine 4.34 · WER 0.0% | Target 0.374 · CE 6.05 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.364 · CE 6.00 · Genuine 4.32 · WER 0.0% | Target 0.381 · CE 6.15 · Genuine 4.00 · WER 0.0% | Target 0.358 · CE 6.16 · Genuine 4.11 · WER 0.0% | Target 0.375 · CE 6.19 · Genuine 4.53 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. 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 Infatuation. 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 Infatuation. 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 0.368 · CE 6.22 · Genuine 4.42 · WER 0.0% | Target 0.363 · CE 6.08 · Genuine 4.46 · WER 0.0% | Target 0.376 · CE 6.13 · Genuine 4.19 · WER 0.0% | Target 0.383 · CE 6.15 · Genuine 4.31 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.367 ± 0.002 | 6.138 ± 0.079 | 4.371 ± 0.140 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.365 ± 0.019 | 6.017 ± 0.024 | 4.302 ± 0.409 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.362 ± 0.005 | 6.080 ± 0.102 | 4.324 ± 0.471 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.366 ± 0.007 | 6.046 ± 0.134 | 4.371 ± 0.271 | 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta. (Intense Infatuation. 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo. | Target 0.365 · CE 6.13 · Genuine 4.24 · WER 0.0% | Target 0.365 · CE 5.98 · Genuine 3.80 · WER 0.0% | Target 0.367 · CE 5.94 · Genuine 3.69 · WER 0.0% | Target 0.357 · CE 6.13 · Genuine 4.18 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta. (Intense Infatuation. 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo. | Target 0.366 · CE 6.25 · Genuine 4.29 · WER 0.0% | Target 0.344 · CE 6.02 · Genuine 4.22 · WER 0.0% | Target 0.362 · CE 6.17 · Genuine 4.28 · WER 0.0% | Target 0.371 · CE 5.97 · Genuine 4.12 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita. (Intense Infatuation. 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 Infatuation. 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 0.368 · CE 6.08 · Genuine 4.39 · WER 0.0% | Target 0.389 · CE 6.04 · Genuine 4.40 · WER 0.0% | Target 0.356 · CE 6.09 · Genuine 4.77 · WER 0.0% | Target 0.374 · CE 5.90 · Genuine 4.47 · 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 Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Infatuation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita. (Intense Infatuation. 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 Infatuation. 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 6.09 · Genuine 4.56 · WER 0.0% | Target 0.360 · CE 6.02 · Genuine 4.78 · WER 0.0% | Target 0.363 · CE 6.13 · Genuine 4.56 · WER 0.0% | Target 0.363 · CE 6.18 · Genuine 4.71 · WER 0.0% |