| LibriTTS clean (test) | | PESQ4.644 | | 67 | 18d ago |
| LibriSpeech (test-clean) | StableCodec | UT MOS4.23 | | 64 | 1mo ago |
| LibriSpeech clean (test) | StableCodec | UTMOS Score4.31 | | 60 | 25d ago |
| LibriTTS (test-other) | CleanCodec@12.5 | UTMOS4.17 | | 57 | 1mo ago |
| AISHELL-2 Chinese | MOSS-Audio-Tokenizer | SIM0.93 | | 54 | 4mo ago |
| LibriSpeech English (test-clean) | MOSS-Audio-Tokenizer | SIM0.97 | | 54 | 4mo ago |
| LibriTTS + ESC-50 clean-100 (test) | FocalSE | PESQ2.97 | | 48 | 18d ago |
| SEED-ZH | MingTok-Audio | PESQ4.21 | | 29 | 1mo ago |
| SeedTTS en (test) | | WER0.0214 | | 21 | 1mo ago |
| Chinese speech | SAC | UTMOS2.99 | | 19 | 4mo ago |
| English speech | WavTokenizer | UTMOS3.92 | | 19 | 4mo ago |
| Salmon Sentiment Consistency emotional 2025b (OOD) | | WER2.9 | | 18 | 4mo ago |
| Seed-TTS English | DashengTokenizer | PESQ4.125 | | 17 | 1mo ago |
| SEED-EN | H-Codec-2.0 (Large) | PESQ2.77 | | 12 | 4mo ago |
| Open Track 2 (test) | Baseline | ScoreQ-ref1.15 | | 12 | 4mo ago |
| Open Track 1 (test) | Baseline | ScoreQ-ref1.36 | | 12 | 4mo ago |
| MLS non-English OOD (700 utterances (7 languages)) | DTM-Codec | UTMOS3.16 | | 10 | 25d ago |
| ReSSInt laryngeal subjects (test) | Masked Multimodal Speech Synthesis Framework | WER40.5 | | 9 | 1mo ago |
| LibriSpeech other (test) | UniAudio-Token | WER6.79 | | 9 | 1mo ago |
| MLS (Multilingual LibriSpeech) Non-English (test) | Mimi-32 | WER7.3 | | 9 | 4mo ago |
| English Read by Japanese accented speech 2007 (OOD) | | WER14.9 | | 9 | 4mo ago |
| Japanese Versatile Speech unseen language speech 2019 (OOD) | | WER4.6 | | 9 | 4mo ago |
| Gigaspeech noisy speech 2021 (OOD) | | WER9.7 | | 9 | 4mo ago |
| LibriSpeech short splits (4s-10s) (test-clean) | AugCodec-3 | WER5.12 | | 8 | 1mo ago |
| Librispeech (test) | MSR-Codec-612 | STOI0.9 | | 8 | 4mo ago |
