Text-to-speech and ASR models ranked by TTS Arena preference, WER, and MOS.
Benchmark data last synced May 7, 2026
| Kind | Compare | |||||
|---|---|---|---|---|---|---|
SS85.9 | asr | 87 | 0.6B | 0.9 GB | ||
SS85.3 | asr | 88 | 0.6B | 0.9 GB | ||
AA83.4 | Alibaba | asr | 87 | 0.6B | 0.9 GB | |
AA74.9 | NVIDIA | asr | 87 | 0.883B | 1.0 GB | |
AA74.7 | Useful Sensors | asr | 87 | 0.245B | 0.7 GB | |
AA74.6 | NVIDIA | asr | 86 | 0.182B | 0.6 GB | |
AA71.9 | NVIDIA | asr | 86 | 0.978B | 1.1 GB | |
AA71.8 | Alibaba | asr | 88 | 1.7B | 1.5 GB | |
BB69.3 | asr | 89 | 2B | 1.7 GB | ||
BB68.9 | Hugging Face | asr | 86 | 0.8B | 1.0 GB | |
BB67.7 | NVIDIA | asr | 85 | 1.1B | 1.2 GB | |
BB67.7 | NVIDIA | asr | 87 | 1B | 1.1 GB | |
BB65.5 | asr | 86 | 1.5B | 1.4 GB | ||
BB65.0 | Mistral AI | asr | 86 | 3B | 2.3 GB | |
BB64.9 | Microsoft | asr | 88 | 5.6B | 3.9 GB |
Model Providers
Every AI lab that ships models we track, ranked by model count, average score, and frontier presence. Compare OpenAI, Anthropic, Meta, Google, Alibaba, and more.
Model Families
Every model series we track, grouped by family. See every Llama, Qwen, GPT, Claude, Gemini, and beyond on a single page.
