| text-to-speech-synthesis-on-open-tts-leaderboard | Fish Audio S2 Pro | #1 | WER: 1.06RTFx: 0.4 |
| text-to-speech-synthesis-on-open-tts-leaderboard | OmniVoice | #2 | WER: 1.1RTFx: 37.56 |
| text-to-speech-synthesis-on-open-tts-leaderboard | Pocket TTS | #3 | WER: 1.11RTFx: 9.09 |
| text-to-speech-synthesis-on-open-tts-leaderboard | Qwen3-TTS 1.7B (CustomVoice) | #4 | WER: 1.17RTFx: 13.32 |
| text-to-speech-synthesis-on-open-tts-leaderboard | CosyVoice 3 0.5B (RL) | #5 | WER: 1.18RTFx: 2.81 |
| text-to-speech-synthesis-on-open-tts-leaderboard | VibeVoice Realtime 0.5B | #6 | WER: 1.3RTFx: 2.71 |
| text-to-speech-synthesis-on-open-tts-leaderboard | SeamlessM4T v2 Large | #7 | WER: 1.57RTFx: 169.5 |
| text-to-speech-synthesis-on-open-tts-leaderboard | SpeechT5 | #8 | WER: 1.79RTFx: 115.99 |
| text-to-speech-synthesis-on-open-tts-leaderboard | MOSS-TTS | #9 | WER: 1.81RTFx: 3.43 |
| text-to-speech-synthesis-on-open-tts-leaderboard | Qwen3-Omni 30B-A3B | #10 | WER: 1.98RTFx: 11.46 |
| text-to-speech-synthesis-on-open-tts-leaderboard | Voxtral 4B TTS | – | WER: 2.17RTFx: 46.84 |
| text-to-speech-synthesis-on-open-tts-leaderboard-all-languages-average | OmniVoice | #1 | WER: 2.72RTFx: 37.52 |
| text-to-speech-synthesis-on-open-tts-leaderboard-all-languages-average | Fish Audio S2 Pro | #2 | WER: 3.16RTFx: 0.4 |
| text-to-speech-synthesis-on-open-tts-leaderboard-all-languages-average | Qwen3-TTS 1.7B (CustomVoice) | #3 | WER: 3.58RTFx: 13.21 |
| text-to-speech-synthesis-on-open-tts-leaderboard-all-languages-average | MOSS-TTS | #4 | WER: 3.77RTFx: 21.5 |
| text-to-speech-synthesis-on-open-tts-leaderboard-all-languages-average | SeamlessM4T v2 Large | #5 | WER: 6.9RTFx: 178.36 |
| text-to-speech-synthesis-on-open-tts-leaderboard-all-languages-average | CosyVoice 3 0.5B (RL) | #6 | WER: 7.48RTFx: 2.73 |
| text-to-speech-synthesis-on-open-tts-leaderboard-all-languages-average | Qwen3-Omni 30B-A3B | #7 | WER: 9.84RTFx: 10.78 |