Benchmark
EmergentTTS-Eval Paralinguistics WER
Scores are percentages. Lower is better.
- Models measured
- 9
- Scores recorded
- 14
- Sources
- 1
- IQ weight
- 1.07
Score distribution
- 23.1%
- Mean
- 20.9%
- Median
- 6.2%
- Std dev
- 18.3%
- 10th percentile
- 31.7%
- 90th percentile
- 17.7%
- Min
This benchmark has a weight of 1.07 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 | Speech 02 HDvoice: English_expressive_narrator· EmergentTTS-Eval data | 17.7% |
| 2 | Gemini 2.5 Flash TTSvoice: Zephyr· EmergentTTS-Eval data | 18.3% |
| 3 | Gemini 2.5 Flash TTSvoice: Zephyr, strong-prompt· EmergentTTS-Eval data | 18.4% |
| 4 | GPT-4o mini Audio Previewvoice: Alloy· EmergentTTS-Eval data | 19% |
| 5 | GPT-4o mini TTSvoice: Alloy, baseline· EmergentTTS-Eval data | 20.6% |
| 6 | GPT-4o Audio Previewvoice: Alloy, strong-prompt· EmergentTTS-Eval data | 20.6% |
| 7 | Gemini 2.5 Pro TTSvoice: Zephyr, strong-prompt· EmergentTTS-Eval data | 20.9% |
| 8 | GPT-4o mini Audio Previewvoice: Alloy, strong-prompt· EmergentTTS-Eval data | 20.9% |
| 9 | GPT-4o Audio Previewvoice: Ballad, strong-prompt· EmergentTTS-Eval data | 21.2% |
| 10 | GPT-4o mini TTSvoice: Alloy, strong-prompt· EmergentTTS-Eval data | 21.3% |
| 11 | GPT-4o Audio Previewvoice: Alloy· EmergentTTS-Eval data | 23.2% |
| 12 | Kokoro 82Mvoice: af_heart· EmergentTTS-Eval data | 28.4% |
| 13 | F5 TTSvoice: basic_ref_en· EmergentTTS-Eval data | 31.7% |
| 14 | Orpheus TTSvoice: Tara· EmergentTTS-Eval data | 40.9% |
Data credits: EmergentTTS-Eval (Boson AI), Apache-2.0 (licence)