Speech Recognition on LibriSpeech Clean
0.0128WERKimi-Audio
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Kimi-Audio2025.12 | 0.0128 | — | |
| Step-Audio2-Mini2025.12 | 0.0133 | — | |
| Audio-FlamingoModel version=32025.12 | 0.0157 | — | |
| Fun-Audio-ChatModel size=30B-A3B2025.12 | 0.0164 | — | |
| Fun-Audio-ChatModel size=8B2025.12 | 0.0171 | — | |
| wav2vec 2.0Backbone=Large, Language Model=None2022.12 | 0.027 | — | |
| WhisperBackbone=Large V2, Evaluation Protocol=Zero-shot2022.12 | 0.027 | 0 | |
| MiMo-Audio2025.12 | 0.0356 | — |