模型排行

来源 Open ASR Leaderboard 更新于: 2026年9月19日
原始来源

排行来源

排名 模型 机构 分数 置信区间 开放状态 证据
1
Edge0/ARK-ASR-3B Average WER ↓
Edge0 4.76 Mixed / see source 未验证 来源
2
OpenMOSS-Team/MOSS-Transcribe-preview-2B Average WER ↓
OpenMOSS 4.87 Mixed / see source 未验证 来源
3
OpenMOSS-Team/MOSS-Transcribe-Diarize Average WER ↓
OpenMOSS 5.17 Mixed / see source 未验证 来源
4
CohereLabs/cohere-transcribe-03-2026 Average WER ↓
Cohere Labs 5.42 Mixed / see source 未验证 来源
5
Edge0/ARK-ASR-0.6B Average WER ↓
Edge0 5.53 Mixed / see source 未验证 来源
6
soundsgoodai/Zipformer-cr-ctc-transducer-XL-290M Average WER ↓
SoundsGoodAI 5.56 Mixed / see source 未验证 来源
7
Qwen/Qwen3-ASR-1.7B Average WER ↓
Qwen 5.76 Mixed / see source 未验证 来源
8
microsoft/Phi-4-multimodal-instruct Average WER ↓
Microsoft 6.02 Mixed / see source 未验证 来源
9
nvidia/parakeet-tdt-0.6b-v2 Average WER ↓
NVIDIA 6.05 Mixed / see source 未验证 来源
10
nvidia/canary-1b-flash Average WER ↓
NVIDIA 6.35 Mixed / see source 未验证 来源
11
kyutai/stt-2.6b-en Average WER ↓
Kyutai 6.40 Mixed / see source 未验证 来源
12
Qwen/Qwen3-ASR-0.6B Average WER ↓
Qwen 6.42 Mixed / see source 未验证 来源
13
nvidia/canary-1b Average WER ↓
NVIDIA 6.50 Mixed / see source 未验证 来源
14
moonshine-ai/moonshine-streaming-medium Average WER ↓
Moonshine AI 6.66 Mixed / see source 未验证 来源
15
soundsgoodai/Zipformer-transducer-XL-290M Average WER ↓
SoundsGoodAI 6.71 Mixed / see source 未验证 来源
16
nvidia/parakeet-tdt-1.1b Average WER ↓
NVIDIA 7.01 Mixed / see source 未验证 来源
17
zai-org/GLM-ASR-Nano-2512 Average WER ↓
Z.ai 7.03 Mixed / see source 未验证 来源
18
Edge0/Audio8-ASR-0.1B Average WER ↓
Edge0 7.03 Mixed / see source 未验证 来源
19
mistralai/Voxtral-Mini-3B-2507 Average WER ↓
Mistral AI_ 7.05 Mixed / see source 未验证 来源
20
nvidia/canary-180m-flash Average WER ↓
NVIDIA 7.12 Mixed / see source 未验证 来源
21
nvidia/parakeet-rnnt-1.1b Average WER ↓
NVIDIA 7.12 Mixed / see source 未验证 来源
22
nvidia/canary-1b-v2 Average WER ↓
NVIDIA 7.15 Mixed / see source 未验证 来源
23
distil-whisper/distil-large-v3.5 Average WER ↓
Whisper Distillation 7.21 Mixed / see source 未验证 来源
24
espnet/owsm_ctc_v4_1B Average WER ↓
ESPnet 7.42 Mixed / see source 未验证 来源
25
openai/whisper-large-v3 Average WER ↓
OpenAI 7.44 Mixed / see source 未验证 来源
26
nvidia/parakeet-tdt_ctc-110m Average WER ↓
NVIDIA 7.49 Mixed / see source 未验证 来源
27
nvidia/parakeet-rnnt-0.6b Average WER ↓
NVIDIA 7.50 Mixed / see source 未验证 来源
28
nvidia/parakeet-ctc-0.6b Average WER ↓
NVIDIA 7.69 Mixed / see source 未验证 来源
29
microsoft/VibeVoice-ASR-HF Average WER ↓
Microsoft 7.77 Mixed / see source 未验证 来源
30
openai/whisper-large-v2 Average WER ↓
OpenAI 7.83 Mixed / see source 未验证 来源
31
moonshine-ai/moonshine-streaming-small Average WER ↓
Moonshine AI 7.84 Mixed / see source 未验证 来源
32
openai/whisper-large Average WER ↓
OpenAI 7.94 Mixed / see source 未验证 来源
33
openai/whisper-medium.en Average WER ↓
OpenAI 8.09 Mixed / see source 未验证 来源
34
espnet/owsm_ctc_v3.1_1B Average WER ↓
ESPnet 8.12 Mixed / see source 未验证 来源
35
nvidia/stt_en_conformer_ctc_large Average WER ↓
NVIDIA 8.32 Mixed / see source 未验证 来源
36
speechbrain/asr-conformer-loquacious Average WER ↓
SpeechBrain 8.48 Mixed / see source 未验证 来源
37
openai/whisper-small.en Average WER ↓
OpenAI 8.59 Mixed / see source 未验证 来源
38
nvidia/stt_en_fastconformer_ctc_large Average WER ↓
NVIDIA 8.96 Mixed / see source 未验证 来源
39
nvidia/stt_en_fastconformer_transducer_large Average WER ↓
NVIDIA 9.06 Mixed / see source 未验证 来源
40
abr-ai/niagara-38m-batch.en Average WER ↓
Applied Brain Research, Inc. 9.36 Mixed / see source 未验证 来源
41
moonshine-ai/moonshine-base Average WER ↓
Moonshine AI 9.99 Mixed / see source 未验证 来源
42
openai/whisper-base.en Average WER ↓
OpenAI 10.32 Mixed / see source 未验证 来源
43
abr-ai/niagara-19m-batch.en Average WER ↓
Applied Brain Research, Inc. 11.01 Mixed / see source 未验证 来源
44
nvidia/stt_en_conformer_ctc_small Average WER ↓
NVIDIA 11.16 Mixed / see source 未验证 来源
45
moonshine-ai/moonshine-streaming-tiny Average WER ↓
Moonshine AI 12 Mixed / see source 未验证 来源
46
moonshine-ai/moonshine-tiny Average WER ↓
Moonshine AI 12.65 Mixed / see source 未验证 来源
47
openai/whisper-tiny.en Average WER ↓
OpenAI 12.81 Mixed / see source 未验证 来源
48
speechbrain/asr-wav2vec2-librispeech Average WER ↓
SpeechBrain 14.35 Mixed / see source 未验证 来源
49
facebook/wav2vec2-large-960h-lv60-self Average WER ↓
AI at Meta 21.27 Mixed / see source 未验证 来源
50
facebook/mms-1b-all Average WER ↓
AI at Meta 22.54 Mixed / see source 未验证 来源