Benchmark
EmergentTTS-Eval Syntactic Complexity Win-Rate
Scores are percentages. Higher is better.
- Models measured
- 9
- Scores recorded
- 13
- Sources
- 1
- IQ weight
- 0.43
Score distribution
- 53.7%
- Mean
- 58.9%
- Median
- 13.4%
- Std dev
- 29.5%
- 10th percentile
- 68%
- 90th percentile
- 25%
- Min
This benchmark has a weight of 0.43 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 Audio Previewvoice: Ballad, strong-prompt· EmergentTTS-Eval data | 68.4% |
| 2 | Gemini 2.5 Flash TTSvoice: Zephyr· EmergentTTS-Eval data | 68% |
| 3 | Gemini 2.5 Flash TTSvoice: Zephyr, strong-prompt· EmergentTTS-Eval data | 67.1% |
| 4 | GPT-4o mini TTSvoice: Alloy, strong-prompt· EmergentTTS-Eval data | 62% |
| 5 | GPT-4o mini Audio Previewvoice: Alloy, strong-prompt· EmergentTTS-Eval data | 60.4% |
| 6 | Gemini 2.5 Pro TTSvoice: Zephyr, strong-prompt· EmergentTTS-Eval data | 60% |
| 7 | GPT-4o Audio Previewvoice: Alloy, strong-prompt· EmergentTTS-Eval data | 58.9% |
| 8 | GPT-4o Audio Previewvoice: Alloy· EmergentTTS-Eval data | 56.8% |
| 9 | GPT-4o mini Audio Previewvoice: Alloy· EmergentTTS-Eval data | 50.2% |
| 10 | Kokoro 82Mvoice: af_heart· EmergentTTS-Eval data | 46.8% |
| 11 | Speech 02 HDvoice: English_expressive_narrator· EmergentTTS-Eval data | 45.5% |
| 12 | Orpheus TTSvoice: Tara· EmergentTTS-Eval data | 29.5% |
| 13 | F5 TTSvoice: basic_ref_en· EmergentTTS-Eval data | 25% |
Data credits: EmergentTTS-Eval (Boson AI), Apache-2.0 (licence)