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: No new configuration passes the held-out gates; retain standard fallback and inspect audio. Bucket scale 0.0, CFG 1.0.
| Held-out condition | n | Target z↑ | Raw head | CE↑ | Genuine↑ | Voice↑ | WER↓ | Valid | QA |
|---|---|---|---|---|---|---|---|---|---|
| standard | 50 | 0.757 | 1.137 | 6.253 | 4.387 | 0.575 | 0.009 | 1.000 | pass |
| manual | 50 | 1.193 | 1.468 | 6.258 | 4.224 | 0.512 | 0.009 | 1.000 | fail |
| adapter | 50 | 1.228 | 1.494 | 6.255 | 4.164 | 0.524 | 0.006 | 1.000 | fail |
| matched_no_adapter | 50 | 1.420 | 1.640 | 6.247 | 4.168 | 0.527 | 0.008 | 1.000 | pass |
Incremental adapter gain versus same-CFG control: -0.192 z;95% cluster interval [-0.44640625701126735, -0.021640613991941705]. Versus best manual-only: 0.035 z. Manual-only versus standard: 0.436 z.
| Scale | CFG | Target z | CE | Genuine | WER | Provisional QA |
|---|---|---|---|---|---|---|
| 0.0 | 1.0 | 1.556 | 6.187 | 4.245 | 0.005 | True |
| 0.5 | 1.0 | 1.518 | 6.201 | 4.218 | 0.000 | False |
| 0.75 | 1.0 | 1.718 | 6.157 | 4.165 | 0.016 | True |
| 1.0 | 1.0 | 1.490 | 6.188 | 4.150 | 0.000 | False |
| 0.0 | 1.25 | 1.630 | 6.151 | 4.228 | 0.009 | False |
| 0.5 | 1.25 | 1.608 | 6.152 | 4.197 | 0.003 | True |
| 0.75 | 1.25 | 1.912 | 6.205 | 4.173 | 0.000 | True |
| 1.0 | 1.25 | 1.756 | 6.157 | 4.124 | 0.003 | False |
| 0.0 | 1.5 | 1.805 | 6.145 | 4.095 | 0.006 | False |
| 0.5 | 1.5 | 1.911 | 6.164 | 4.173 | 0.009 | False |
| 0.75 | 1.5 | 1.804 | 6.175 | 4.131 | 0.006 | False |
| 1.0 | 1.5 | 1.968 | 6.171 | 4.141 | 0.000 | False |
| 0.0 | 2.0 | 2.124 | 6.146 | 4.117 | 0.005 | False |
| 0.5 | 2.0 | 2.035 | 6.086 | 4.050 | 0.003 | False |
| 0.75 | 2.0 | 2.263 | 6.113 | 4.056 | 0.000 | False |
| 1.0 | 2.0 | 2.281 | 6.047 | 3.999 | 0.000 | False |
| 0.0 | 2.5 | 2.432 | 6.004 | 3.971 | 0.030 | False |
| 0.5 | 2.5 | 2.457 | 6.046 | 3.975 | 0.005 | False |
| 0.75 | 2.5 | 2.561 | 6.025 | 3.973 | 0.012 | False |
| 1.0 | 2.5 | 2.542 | 6.087 | 4.001 | 0.000 | False |
| 0.0 | 3.0 | 2.670 | 5.986 | 3.968 | 0.000 | False |
| 0.5 | 3.0 | 2.768 | 5.979 | 3.959 | 0.003 | False |
| 0.75 | 3.0 | 2.591 | 5.963 | 3.955 | 0.000 | False |
| 1.0 | 3.0 | 2.628 | 5.980 | 3.973 | 0.017 | False |
| 0.0 | 4.0 | 2.942 | 5.838 | 3.940 | 0.008 | False |
| 0.5 | 4.0 | 2.991 | 5.819 | 3.884 | 0.003 | False |
| 0.75 | 4.0 | 2.906 | 5.876 | 3.904 | 0.021 | False |
| 1.0 | 4.0 | 2.878 | 5.897 | 3.910 | 0.008 | False |
| 0.0 | 5.0 | 2.945 | 5.859 | 3.898 | 0.022 | False |
| 0.5 | 5.0 | 3.087 | 5.757 | 3.881 | 0.048 | False |
| 0.75 | 5.0 | 3.060 | 5.786 | 3.855 | 0.005 | False |
| 1.0 | 5.0 | 3.080 | 5.724 | 3.865 | 0.009 | False |
direct_manual_recipe
{
"id": "manual-d1f4f47e8466",
"general": "Channel an eager, energetic anticipation, sounding like someone eagerly awaiting a wonderful outcome.",
"inline": [
"Initiate with a slightly higher vocal register, conveying immediate excitement and forward motion.",
"Keep the pace lively and conversational, but maintain a focused, positive intent throughout.",
"Sustain the energetic quality right to the final word, making the listener feel the momentum."
],
"cfg": 3,
"recipe": "P07",
"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 Hope Enthusiasm Optimism. 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": 19, "nl": 12, "de": 32, "en": 27, "fr": 11}. 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 | 0.804 ± 0.161 | 6.215 ± 0.119 | 4.372 ± 0.278 | 0.83% ± 3.73 |
| Rank32 × 0.5 | 20 | 0.805 ± 0.160 | 6.193 ± 0.099 | 4.383 ± 0.271 | 0.28% ± 1.24 |
| Rank32 × 0.75 | 20 | 0.816 ± 0.153 | 6.201 ± 0.115 | 4.357 ± 0.203 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 20 | 0.782 ± 0.183 | 6.212 ± 0.124 | 4.389 ± 0.232 | 0.31% ± 1.40 |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.936 ± 0.032 | 6.330 ± 0.041 | 4.208 ± 0.326 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.900 ± 0.044 | 6.239 ± 0.126 | 4.135 ± 0.323 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.901 ± 0.084 | 6.269 ± 0.043 | 4.198 ± 0.295 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.930 ± 0.031 | 6.311 ± 0.040 | 4.269 ± 0.233 | 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.926 · CE 6.32 · Genuine 3.90 · WER 0.0% | Target 0.856 · CE 6.17 · Genuine 3.98 · WER 0.0% | Target 0.777 · CE 6.27 · Genuine 4.18 · WER 0.0% | Target 0.894 · CE 6.28 · Genuine 3.99 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.897 · CE 6.32 · Genuine 4.13 · WER 0.0% | Target 0.870 · CE 6.09 · Genuine 3.80 · WER 0.0% | Target 0.943 · CE 6.21 · Genuine 3.80 · WER 0.0% | Target 0.915 · CE 6.27 · Genuine 4.17 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.946 · CE 6.39 · Genuine 4.67 · WER 0.0% | Target 0.930 · CE 6.34 · Genuine 4.55 · WER 0.0% | Target 0.921 · CE 6.28 · Genuine 4.30 · WER 0.0% | Target 0.943 · CE 6.34 · Genuine 4.51 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.974 · CE 6.29 · Genuine 4.12 · WER 0.0% | Target 0.945 · CE 6.35 · Genuine 4.22 · WER 0.0% | Target 0.962 · CE 6.32 · Genuine 4.51 · WER 0.0% | Target 0.966 · CE 6.35 · Genuine 4.41 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.815 ± 0.157 | 6.261 ± 0.120 | 4.307 ± 0.232 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.866 ± 0.099 | 6.287 ± 0.127 | 4.410 ± 0.220 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.825 ± 0.168 | 6.309 ± 0.103 | 4.424 ± 0.139 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.779 ± 0.210 | 6.311 ± 0.122 | 4.289 ± 0.126 | 1.56% ± 3.12 |
| 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.766 · CE 6.24 · Genuine 4.39 · WER 0.0% | Target 0.838 · CE 6.25 · Genuine 4.23 · WER 0.0% | Target 0.626 · CE 6.17 · Genuine 4.37 · WER 0.0% | Target 0.573 · CE 6.13 · Genuine 4.34 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.615 · CE 6.14 · Genuine 4.60 · WER 0.0% | Target 0.739 · CE 6.13 · Genuine 4.72 · WER 0.0% | Target 0.747 · CE 6.28 · Genuine 4.48 · WER 0.0% | Target 0.623 · CE 6.32 · Genuine 4.44 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen. | Target 0.927 · CE 6.43 · Genuine 4.16 · WER 0.0% | Target 0.954 · CE 6.38 · Genuine 4.41 · WER 0.0% | Target 0.968 · CE 6.40 · Genuine 4.26 · WER 0.0% | Target 0.964 · CE 6.40 · Genuine 4.17 · WER 6.2% |
| 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen. | Target 0.953 · CE 6.24 · Genuine 4.09 · WER 0.0% | Target 0.934 · CE 6.40 · Genuine 4.28 · WER 0.0% | Target 0.961 · CE 6.38 · Genuine 4.58 · WER 0.0% | Target 0.954 · CE 6.39 · Genuine 4.20 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.867 ± 0.130 | 6.243 ± 0.099 | 4.584 ± 0.294 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.900 ± 0.093 | 6.144 ± 0.012 | 4.561 ± 0.240 | 1.39% ± 2.78 |
| Rank32 × 0.75 | 4 | 0.921 ± 0.079 | 6.203 ± 0.074 | 4.564 ± 0.149 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.865 ± 0.127 | 6.249 ± 0.033 | 4.708 ± 0.138 | 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.777 · CE 6.22 · Genuine 4.58 · WER 0.0% | Target 0.882 · CE 6.15 · Genuine 4.49 · WER 5.6% | Target 0.805 · CE 6.17 · Genuine 4.57 · WER 0.0% | Target 0.726 · CE 6.25 · Genuine 4.68 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.735 · CE 6.14 · Genuine 4.17 · WER 0.0% | Target 0.776 · CE 6.16 · Genuine 4.26 · WER 0.0% | Target 0.948 · CE 6.25 · Genuine 4.37 · WER 0.0% | Target 0.789 · CE 6.21 · Genuine 4.76 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.973 · CE 6.23 · Genuine 4.73 · WER 0.0% | Target 0.975 · CE 6.13 · Genuine 4.68 · WER 0.0% | Target 0.954 · CE 6.12 · Genuine 4.58 · WER 0.0% | Target 0.966 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.984 · CE 6.38 · Genuine 4.85 · WER 0.0% | Target 0.967 · CE 6.13 · Genuine 4.81 · WER 0.0% | Target 0.979 · CE 6.28 · Genuine 4.73 · WER 0.0% | Target 0.980 · CE 6.26 · Genuine 4.86 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.617 ± 0.131 | 6.166 ± 0.088 | 4.324 ± 0.134 | 4.17% ± 8.33 |
| Rank32 × 0.5 | 4 | 0.586 ± 0.100 | 6.152 ± 0.062 | 4.343 ± 0.036 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.601 ± 0.084 | 6.126 ± 0.142 | 4.328 ± 0.103 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.615 ± 0.123 | 6.120 ± 0.091 | 4.300 ± 0.099 | 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.530 · CE 6.10 · Genuine 4.15 · WER 0.0% | Target 0.640 · CE 6.14 · Genuine 4.31 · WER 0.0% | Target 0.492 · CE 5.97 · Genuine 4.28 · WER 0.0% | Target 0.693 · CE 6.02 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.486 · CE 6.08 · Genuine 4.39 · WER 0.0% | Target 0.623 · CE 6.16 · Genuine 4.36 · WER 0.0% | Target 0.689 · CE 6.23 · Genuine 4.28 · WER 0.0% | Target 0.449 · CE 6.06 · Genuine 4.23 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.685 · CE 6.23 · Genuine 4.29 · WER 0.0% | Target 0.436 · CE 6.08 · Genuine 4.32 · WER 0.0% | Target 0.636 · CE 6.05 · Genuine 4.27 · WER 0.0% | Target 0.722 · CE 6.20 · Genuine 4.32 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.766 · CE 6.25 · Genuine 4.46 · WER 16.7% | Target 0.643 · CE 6.23 · Genuine 4.38 · WER 0.0% | Target 0.585 · CE 6.26 · Genuine 4.48 · WER 0.0% | Target 0.594 · CE 6.20 · Genuine 4.43 · WER 0.0% |
Fixed speaker reference:
| Condition | n | Target score ↑ | Content Enjoyment ↑ | Genuineness ↑ | WER ↓ |
|---|---|---|---|---|---|
| Merged base, no bucket adapter | 4 | 0.785 ± 0.170 | 6.074 ± 0.066 | 4.437 ± 0.337 | 0.00% ± 0.00 |
| Rank32 × 0.5 | 4 | 0.774 ± 0.194 | 6.146 ± 0.060 | 4.468 ± 0.342 | 0.00% ± 0.00 |
| Rank32 × 0.75 | 4 | 0.831 ± 0.111 | 6.100 ± 0.055 | 4.272 ± 0.115 | 0.00% ± 0.00 |
| Rank32 × 1.0 | 4 | 0.720 ± 0.234 | 6.070 ± 0.080 | 4.379 ± 0.252 | 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta. (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo. | Target 0.626 · CE 5.98 · Genuine 4.26 · WER 0.0% | Target 0.542 · CE 6.11 · Genuine 4.35 · WER 0.0% | Target 0.673 · CE 6.18 · Genuine 4.28 · WER 0.0% | Target 0.672 · CE 6.01 · Genuine 4.01 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta. (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo. | Target 0.651 · CE 6.12 · Genuine 4.06 · WER 0.0% | Target 0.693 · CE 6.09 · Genuine 4.04 · WER 0.0% | Target 0.852 · CE 6.08 · Genuine 4.32 · WER 0.0% | Target 0.410 · CE 6.09 · Genuine 4.46 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita. (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.934 · CE 6.13 · Genuine 4.63 · WER 0.0% | Target 0.971 · CE 6.22 · Genuine 4.67 · WER 0.0% | Target 0.929 · CE 6.06 · Genuine 4.38 · WER 0.0% | Target 0.866 · CE 6.17 · Genuine 4.55 · 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 Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word. SCRIPT: (Intense Hope Enthusiasm Optimism. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita. (Intense Hope Enthusiasm Optimism. 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 Hope Enthusiasm Optimism. 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 0.930 · CE 6.07 · Genuine 4.80 · WER 0.0% | Target 0.890 · CE 6.17 · Genuine 4.81 · WER 0.0% | Target 0.870 · CE 6.08 · Genuine 4.11 · WER 0.0% | Target 0.931 · CE 6.00 · Genuine 4.51 · WER 0.0% |