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 1.0.
| Held-out condition | n | Target z↑ | Raw head | CE↑ | Genuine↑ | Voice↑ | WER↓ | Valid | QA |
|---|---|---|---|---|---|---|---|---|---|
| standard | 50 | -0.404 | 0.162 | 6.260 | 4.393 | 0.576 | 0.006 | 1.000 | pass |
| manual | 50 | -0.106 | 0.376 | 6.278 | 4.298 | 0.577 | 0.008 | 1.000 | pass |
| adapter | 50 | 0.869 | 1.076 | 6.206 | 4.126 | 0.535 | 0.023 | 0.980 | pass |
| matched_no_adapter | 50 | 1.024 | 1.188 | 6.217 | 4.145 | 0.521 | 0.034 | 0.960 | fail |
Incremental adapter gain versus same-CFG control: -0.155 z;95% cluster interval [-0.3934374861039119, 0.062265604104016714]. Versus best manual-only: 0.975 z. Manual-only versus standard: 0.299 z.
| Scale | CFG | Target z | CE | Genuine | WER | Provisional QA |
|---|---|---|---|---|---|---|
| 0.0 | 1.0 | -0.019 | 6.207 | 4.171 | 0.000 | True |
| 0.5 | 1.0 | 0.217 | 6.190 | 4.283 | 0.007 | False |
| 0.75 | 1.0 | 0.123 | 6.189 | 4.222 | 0.000 | True |
| 1.0 | 1.0 | 0.111 | 6.224 | 4.283 | 0.003 | True |
| 0.0 | 1.25 | 0.391 | 6.203 | 4.199 | 0.000 | False |
| 0.5 | 1.25 | 0.409 | 6.179 | 4.231 | 0.003 | True |
| 0.75 | 1.25 | 0.386 | 6.185 | 4.225 | 0.000 | True |
| 1.0 | 1.25 | 0.367 | 6.170 | 4.208 | 0.003 | True |
| 0.0 | 1.5 | 0.648 | 6.161 | 4.189 | 0.003 | False |
| 0.5 | 1.5 | 0.549 | 6.154 | 4.134 | 0.000 | False |
| 0.75 | 1.5 | 0.592 | 6.161 | 4.189 | 0.000 | True |
| 1.0 | 1.5 | 0.533 | 6.169 | 4.143 | 0.003 | False |
| 0.0 | 2.0 | 0.888 | 6.147 | 4.119 | 0.012 | False |
| 0.5 | 2.0 | 0.813 | 6.178 | 4.123 | 0.013 | False |
| 0.75 | 2.0 | 0.767 | 6.135 | 4.110 | 0.025 | False |
| 1.0 | 2.0 | 0.873 | 6.138 | 4.148 | 0.006 | False |
| 0.0 | 2.5 | 1.415 | 6.093 | 4.039 | 0.012 | False |
| 0.5 | 2.5 | 1.235 | 6.113 | 4.052 | 0.009 | False |
| 0.75 | 2.5 | 1.455 | 6.082 | 4.022 | 0.021 | False |
| 1.0 | 2.5 | 1.331 | 6.102 | 3.963 | 0.006 | True |
| 0.0 | 3.0 | 1.677 | 6.030 | 3.968 | 0.006 | False |
| 0.5 | 3.0 | 1.682 | 6.026 | 4.013 | 0.021 | False |
| 0.75 | 3.0 | 1.767 | 6.003 | 3.985 | 0.006 | False |
| 1.0 | 3.0 | 1.713 | 6.052 | 3.985 | 0.009 | False |
| 0.0 | 4.0 | 2.174 | 5.902 | 3.900 | 0.009 | False |
| 0.5 | 4.0 | 2.192 | 5.880 | 3.895 | 0.021 | False |
| 0.75 | 4.0 | 2.373 | 5.787 | 3.857 | 0.036 | False |
| 1.0 | 4.0 | 2.488 | 5.835 | 3.870 | 0.015 | False |
| 0.0 | 5.0 | 2.827 | 5.685 | 3.858 | 0.014 | False |
| 0.5 | 5.0 | 2.351 | 5.651 | 3.834 | 0.039 | False |
| 0.75 | 5.0 | 2.915 | 5.639 | 3.843 | 0.025 | False |
| 1.0 | 5.0 | 2.782 | 5.736 | 3.833 | 0.058 | False |
direct_manual_recipe
{
"id": "manual-a8a332cdd043",
"general": "Adopt a grand, expansive vocal projection, as if sharing monumental good news with the world.",
"inline": [
"Project the initial sounds with full chest resonance, conveying overwhelming, open-hearted gladness.",
"Expand the vocal space during the middle, letting the sound swell with expansive, unrestrained glee.",
"Deliver the final words with a powerful, sustained upward thrust, a triumphant vocal peak."
],
"cfg": 3,
"recipe": "P02",
"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 Elation. 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: {"de": 26, "en": 24, "fr": 3, "nl": 3, "es": 9}. 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 | 0.366 ± 0.110 | 6.204 ± 0.113 | 4.343 ± 0.254 | 0.83% ± 3.73 |
| Rank32 × 0.5 | 20 | 0.349 ± 0.099 | 6.191 ± 0.122 | 4.356 ± 0.187 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 20 | 0.386 ± 0.128 | 6.176 ± 0.123 | 4.333 ± 0.241 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 20 | 0.348 ± 0.083 | 6.189 ± 0.127 | 4.317 ± 0.215 | 0.83% ± 3.73 |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.444 ± 0.220 | 6.239 ± 0.121 | 4.153 ± 0.272 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.389 ± 0.180 | 6.301 ± 0.085 | 4.272 ± 0.215 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.432 ± 0.208 | 6.269 ± 0.101 | 4.103 ± 0.193 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.379 ± 0.164 | 6.281 ± 0.108 | 4.183 ± 0.291 | 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.266 · CE 6.13 · Genuine 3.98 · WER 0.0% | Target 0.236 · CE 6.30 · Genuine 4.18 · WER 0.0% | Target 0.241 · CE 6.26 · Genuine 3.94 · WER 0.0% | Target 0.266 · CE 6.25 · Genuine 4.08 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.255 · CE 6.14 · Genuine 3.87 · WER 0.0% | Target 0.311 · CE 6.21 · Genuine 4.09 · WER 0.0% | Target 0.262 · CE 6.14 · Genuine 3.94 · WER 0.0% | Target 0.287 · CE 6.14 · Genuine 3.82 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.697 · CE 6.38 · Genuine 4.33 · WER 0.0% | Target 0.648 · CE 6.28 · Genuine 4.23 · WER 0.0% | Target 0.608 · CE 6.39 · Genuine 4.21 · WER 0.0% | Target 0.620 · CE 6.39 · Genuine 4.41 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.560 · CE 6.30 · Genuine 4.43 · WER 0.0% | Target 0.363 · CE 6.41 · Genuine 4.58 · WER 0.0% | Target 0.617 · CE 6.28 · Genuine 4.32 · WER 0.0% | Target 0.345 · CE 6.34 · Genuine 4.43 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.406 ± 0.072 | 6.297 ± 0.019 | 4.394 ± 0.200 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.389 ± 0.125 | 6.267 ± 0.134 | 4.462 ± 0.130 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.399 ± 0.157 | 6.257 ± 0.123 | 4.523 ± 0.131 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.364 ± 0.051 | 6.332 ± 0.142 | 4.307 ± 0.026 | 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.370 · CE 6.28 · Genuine 4.36 · WER 0.0% | Target 0.299 · CE 6.23 · Genuine 4.58 · WER 0.0% | Target 0.332 · CE 6.19 · Genuine 4.60 · WER 0.0% | Target 0.317 · CE 6.21 · Genuine 4.32 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.322 · CE 6.28 · Genuine 4.67 · WER 0.0% | Target 0.304 · CE 6.09 · Genuine 4.31 · WER 0.0% | Target 0.307 · CE 6.17 · Genuine 4.41 · WER 0.0% | Target 0.325 · CE 6.21 · Genuine 4.27 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen. | Target 0.465 · CE 6.32 · Genuine 4.19 · WER 0.0% | Target 0.566 · CE 6.36 · Genuine 4.56 · WER 0.0% | Target 0.634 · CE 6.24 · Genuine 4.42 · WER 0.0% | Target 0.398 · CE 6.46 · Genuine 4.31 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen. | Target 0.467 · CE 6.31 · Genuine 4.36 · WER 0.0% | Target 0.386 · CE 6.38 · Genuine 4.39 · WER 0.0% | Target 0.321 · CE 6.44 · Genuine 4.67 · WER 0.0% | Target 0.418 · CE 6.45 · Genuine 4.33 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.288 ± 0.011 | 6.189 ± 0.097 | 4.569 ± 0.169 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.313 ± 0.068 | 6.150 ± 0.070 | 4.385 ± 0.126 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.308 ± 0.020 | 6.181 ± 0.043 | 4.357 ± 0.346 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.290 ± 0.063 | 6.138 ± 0.049 | 4.510 ± 0.203 | 4.17% ± 8.33 |
| 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.295 · CE 6.18 · Genuine 4.53 · WER 0.0% | Target 0.261 · CE 6.14 · Genuine 4.39 · WER 0.0% | Target 0.296 · CE 6.16 · Genuine 4.36 · WER 0.0% | Target 0.247 · CE 6.08 · Genuine 4.47 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.280 · CE 6.16 · Genuine 4.41 · WER 0.0% | Target 0.251 · CE 6.05 · Genuine 4.24 · WER 0.0% | Target 0.291 · CE 6.16 · Genuine 3.90 · WER 0.0% | Target 0.269 · CE 6.14 · Genuine 4.39 · WER 16.7% |
| 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.278 · CE 6.32 · Genuine 4.81 · WER 0.0% | Target 0.388 · CE 6.20 · Genuine 4.36 · WER 0.0% | Target 0.310 · CE 6.25 · Genuine 4.42 · WER 0.0% | Target 0.384 · CE 6.13 · Genuine 4.37 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.300 · CE 6.09 · Genuine 4.52 · WER 0.0% | Target 0.354 · CE 6.21 · Genuine 4.55 · WER 0.0% | Target 0.335 · CE 6.17 · Genuine 4.75 · WER 0.0% | Target 0.262 · CE 6.20 · Genuine 4.81 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.358 ± 0.043 | 6.237 ± 0.084 | 4.385 ± 0.072 | 4.17% ± 8.33 |
| Rank32 × 0.5 | 4 | 0.335 ± 0.027 | 6.144 ± 0.104 | 4.313 ± 0.183 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.372 ± 0.031 | 6.136 ± 0.135 | 4.289 ± 0.088 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.377 ± 0.039 | 6.084 ± 0.043 | 4.299 ± 0.158 | 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.402 · CE 6.22 · Genuine 4.28 · WER 0.0% | Target 0.375 · CE 6.02 · Genuine 4.04 · WER 0.0% | Target 0.350 · CE 6.08 · Genuine 4.34 · WER 0.0% | Target 0.345 · CE 6.12 · Genuine 4.22 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.314 · CE 6.16 · Genuine 4.42 · WER 0.0% | Target 0.322 · CE 6.14 · Genuine 4.36 · WER 0.0% | Target 0.349 · CE 6.00 · Genuine 4.28 · WER 0.0% | Target 0.355 · CE 6.02 · Genuine 4.33 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.387 · CE 6.21 · Genuine 4.39 · WER 0.0% | Target 0.330 · CE 6.13 · Genuine 4.43 · WER 0.0% | Target 0.376 · CE 6.15 · Genuine 4.17 · WER 0.0% | Target 0.376 · CE 6.11 · Genuine 4.51 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. 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 Elation. 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 Elation. 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 0.329 · CE 6.36 · Genuine 4.44 · WER 16.7% | Target 0.314 · CE 6.28 · Genuine 4.42 · WER 0.0% | Target 0.414 · CE 6.32 · Genuine 4.36 · WER 0.0% | Target 0.432 · CE 6.09 · Genuine 4.14 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.336 ± 0.045 | 6.056 ± 0.068 | 4.212 ± 0.340 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.316 ± 0.036 | 6.092 ± 0.104 | 4.349 ± 0.283 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.422 ± 0.144 | 6.035 ± 0.038 | 4.394 ± 0.240 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.327 ± 0.036 | 6.108 ± 0.029 | 4.285 ± 0.256 | 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta. (Intense Elation. 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo. | Target 0.303 · CE 6.10 · Genuine 4.61 · WER 0.0% | Target 0.269 · CE 6.05 · Genuine 3.94 · WER 0.0% | Target 0.307 · CE 5.98 · Genuine 4.09 · WER 0.0% | Target 0.298 · CE 6.13 · Genuine 4.13 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta. (Intense Elation. 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo. | Target 0.302 · CE 5.98 · Genuine 3.79 · WER 0.0% | Target 0.311 · CE 5.97 · Genuine 4.58 · WER 0.0% | Target 0.309 · CE 6.05 · Genuine 4.36 · WER 0.0% | Target 0.295 · CE 6.07 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita. (Intense Elation. 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 Elation. 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 0.397 · CE 6.13 · Genuine 4.14 · WER 0.0% | Target 0.348 · CE 6.12 · Genuine 4.43 · WER 0.0% | Target 0.608 · CE 6.07 · Genuine 4.46 · WER 0.0% | Target 0.368 · CE 6.13 · Genuine 4.52 · 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 Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Elation. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita. (Intense Elation. 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 Elation. 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 6.01 · Genuine 4.31 · WER 0.0% | Target 0.339 · CE 6.22 · Genuine 4.44 · WER 0.0% | Target 0.463 · CE 6.03 · Genuine 4.67 · WER 0.0% | Target 0.347 · CE 6.10 · Genuine 4.48 · WER 0.0% |