Benchmark
EmergentTTS-Eval Foreign Words Win-Rate
Scores are percentages. Higher is better.
- Models measured
- 9
- Scores recorded
- 13
- Sources
- 1
- IQ weight
- 0.32
Score distribution
- 51.3%
- Mean
- 63.4%
- Median
- 24.6%
- Std dev
- 13.2%
- 10th percentile
- 71.6%
- 90th percentile
- 3.2%
- Min
This benchmark has a weight of 0.32 for speech models, measured over 7 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o mini Audio Previewvoice: Alloy, strong-prompt· EmergentTTS-Eval data | 73.4% |
| 2 | GPT-4o Audio Previewvoice: Alloy, strong-prompt· EmergentTTS-Eval data | 71.6% |
| 3 | GPT-4o Audio Previewvoice: Ballad, strong-prompt· EmergentTTS-Eval data | 70% |
| 4 | GPT-4o Audio Previewvoice: Alloy· EmergentTTS-Eval data | 68.8% |
| 5 | Gemini 2.5 Flash TTSvoice: Zephyr, strong-prompt· EmergentTTS-Eval data | 65.6% |
| 6 | GPT-4o mini Audio Previewvoice: Alloy· EmergentTTS-Eval data | 64.8% |
| 7 | Gemini 2.5 Pro TTSvoice: Zephyr, strong-prompt· EmergentTTS-Eval data | 63.4% |
| 8 | Gemini 2.5 Flash TTSvoice: Zephyr· EmergentTTS-Eval data | 63% |
| 9 | GPT-4o mini TTSvoice: Alloy, strong-prompt· EmergentTTS-Eval data | 63% |
| 10 | Speech 02 HDvoice: English_expressive_narrator· EmergentTTS-Eval data | 32% |
| 11 | Orpheus TTSvoice: Tara· EmergentTTS-Eval data | 14.6% |
| 12 | Kokoro 82Mvoice: af_heart· EmergentTTS-Eval data | 13.2% |
| 13 | F5 TTSvoice: basic_ref_en· EmergentTTS-Eval data | 3.2% |
Data credits: EmergentTTS-Eval (Boson AI), Apache-2.0 (licence)