Codesota · Speech · Speech Recognition · Open ASR LeaderboardTasks/Speech/Speech Recognition
Speech Recognition · benchmark dataset · 2023 · EN

HF Open ASR Leaderboard (aggregate).

The Hugging Face Open ASR Leaderboard aggregates Word Error Rate and real-time factor across LibriSpeech, AMI, Earnings-22, GigaSpeech, SPGISpeech, TED-LIUM, and VoxPopuli to give a single composite score for English ASR systems. The de-facto modern ASR leaderboard.

Paper ↗Submit a result ↵
§ 01 · Leaderboard

Best published scores.

102 results indexed across 2 metrics. Shaded row marks current SOTA; ties broken by submission date.

Primary
wer · lower is better
All metrics
rtfx, wer
rtfx
50 rows
#ModelOrgSubmittedPaper / codertfx
01Stt_en_fastconformer_ctc_large—May 2023Fast Conformer with Linearly Scalable Attention for Effi…6399.25
02Stt_en_conformer_ctc_small—May 2020Conformer: Convolution-augmented Transformer for Speech … · code5686.90
03Parakeet-tdt_ctc-110m—Apr 2023Efficient Sequence Transduction by Jointly Predicting To… · code5345.14
04Stt_en_conformer_ctc_large—May 2020Conformer: Convolution-augmented Transformer for Speech … · code4295.01
05Parakeet-ctc-0.6b—May 2023Fast Conformer with Linearly Scalable Attention for Effi…4281.53
06Stt_en_fastconformer_transducer_large—May 2023Fast Conformer with Linearly Scalable Attention for Effi…4097.43
07Parakeet-tdt-0.6b-v2—Apr 2023Efficient Sequence Transduction by Jointly Predicting To… · code3386.02
08Parakeet-rnnt-0.6b—May 2023Fast Conformer with Linearly Scalable Attention for Effi…2815.72
09Moonshine-streaming-tiny—Jan 2026pwc-dump847.20
10Moonshine-tiny—Oct 2024Moonshine: Speech Recognition for Live Transcription and… · code753.06
11Wav2vec2-base-960h—Jun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of… · code686.00
12Data2vec-audio-base-960h—Feb 2022data2vec: A General Framework for Self-supervised Learni… · code648.14
13Wav2vec2-conformer-rope-large-960h-ft—Oct 2020fairseq S2T: Fast Speech-to-Text Modeling with fairseq · code607.87
14Moonshine-streaming-small—Jan 2026pwc-dump566.33
15Moonshine-base—Oct 2024Moonshine: Speech Recognition for Live Transcription and… · code565.97
16Cohere Transcribe (Mar 2026)OpenCohereMar 2026pwc-dump524.88
17Wav2vec2-conformer-rel-pos-large-960h-ft—Oct 2020fairseq S2T: Fast Speech-to-Text Modeling with fairseq · code522.46
18wav2vec 2.0 Large (960h)OpenMeta AIJun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of… · code516.58
19Wav2vec2-large-960h-lv60-self—Jun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of… · code509.32
20Wav2vec2-large-robust-ft-libri-960h—Apr 2021Robust wav2vec 2.0: Analyzing Domain Shift in Self-Super… · code503.81
21Owsm_ctc_v3.1_1B—Jan 2024OWSM v3.1: Better and Faster Open Whisper-Style Speech M… · code502.02
22Hubert-large-ls960-ft—Jun 2021HuBERT: Self-Supervised Speech Representation Learning b… · code495.86
23Data2vec-audio-large-960h—Feb 2022data2vec: A General Framework for Self-supervised Learni… · code470.15
24Asr-wav2vec2-librispeech—Jun 2021SpeechBrain: A General-Purpose Speech Toolkit · code451.18
25Moonshine Streaming MediumOpenUseful SensorsJan 2026pwc-dump448.15
26Hubert-xlarge-ls960-ft—Jun 2021HuBERT: Self-Supervised Speech Representation Learning b… · code361.32
27Whisper-tiny.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code348.12
28Distil-small.en—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code331.89
29Whisper-base.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code320.67
30Distil-medium.en—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code279.73
31Granite Speech 3.3 2BOpenIBMMay 2025Granite-speech: open-source speech-aware LLMs with stron…270.57
32Whisper-small.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code268.91
33Mms-1b-fl102—May 2023Scaling Speech Technology to 1,000+ Languages · code234.42
34Granite Speech 4.1 2BOpenIBMMay 2025Granite-speech: open-source speech-aware LLMs with stron…231.29
35Mms-1b-all—May 2023Scaling Speech Technology to 1,000+ Languages · code230.79
36Distil-large-v3—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code214.42
37Distil-large-v2—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code202.95
38Whisper Large v3 TurboOpenOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code200.19
39Lite-whisper-large-v3-turbo-acc—Feb 2025LiteASR: Efficient Automatic Speech Recognition with Low… · code191.71
40Whisper-medium.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code182.13
41Phi-4 Multimodal InstructOpenMicrosoftMar 2025Phi-4-Mini Technical Report: Compact yet Powerful Multim…151.10
42Qwen3-ASR-1.7BOpenAlibabaJan 2026Qwen3-ASR Technical Report · code147.93
43Whisper Large v3OpenOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code145.51
44Whisper Large v2OpenOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code144.45
45Whisper Large—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code143.76
46Lite-whisper-large-v3-fast—Feb 2025LiteASR: Efficient Automatic Speech Recognition with Low… · code120.76
47Voxtral-Mini-4B-Realtime-2602OpenMistral AIFeb 2026Voxtral Realtime93.32
48SYMPHONY-ASR—Jan 2026pwc-dump77.56
49VibeVoice-ASR-HF—Jan 2026VIBEVOICE-ASR Technical Report51.80
50Asr-conformer-loquacious—Feb 2025pwc-dump42.16
wer· primary
52 rows
#ModelOrgSubmittedPaper / codewer
01Granite Speech 4.1 2BOpenIBMMay 2025Granite-speech: open-source speech-aware LLMs with stron…5.33
02Cohere Transcribe (Mar 2026)OpenCohereMar 2026pwc-dump5.42
03Qwen3-ASR-1.7BOpenAlibabaJan 2026Qwen3-ASR Technical Report · code5.76
04SYMPHONY-ASR—Jan 2026pwc-dump5.91
05Granite Speech 3.3 2BOpenIBMMay 2025Granite-speech: open-source speech-aware LLMs with stron…6.00
06Phi-4 Multimodal InstructOpenMicrosoftMar 2025Phi-4-Mini Technical Report: Compact yet Powerful Multim…6.02
07Parakeet-tdt-0.6b-v2—Apr 2023Efficient Sequence Transduction by Jointly Predicting To… · code6.05
08Moonshine Streaming MediumOpenUseful SensorsJan 2026pwc-dump6.66
09Whisper Large v3OpenOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code7.44
10Parakeet-tdt_ctc-110m—Apr 2023Efficient Sequence Transduction by Jointly Predicting To… · code7.49
11Parakeet-rnnt-0.6b—May 2023Fast Conformer with Linearly Scalable Attention for Effi…7.50
12Distil-large-v3—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code7.52
13Voxtral-Mini-4B-Realtime-2602OpenMistral AIFeb 2026Voxtral Realtime7.68
14Parakeet-ctc-0.6b—May 2023Fast Conformer with Linearly Scalable Attention for Effi…7.69
15Lite-whisper-large-v3-turbo-acc—Feb 2025LiteASR: Efficient Automatic Speech Recognition with Low… · code7.77
16VibeVoice-ASR-HF—Jan 2026VIBEVOICE-ASR Technical Report7.77
17Whisper Large v3 TurboOpenOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code7.83
18Whisper Large v2OpenOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code7.83
19Moonshine-streaming-small—Jan 2026pwc-dump7.84
20Distil-large-v2—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code7.92
21Whisper Large—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code7.94
22Whisper-medium.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code8.09
23Owsm_ctc_v3.1_1B—Jan 2024OWSM v3.1: Better and Faster Open Whisper-Style Speech M… · code8.12
24Lite-whisper-large-v3-fast—Feb 2025LiteASR: Efficient Automatic Speech Recognition with Low… · code8.16
25Stt_en_conformer_ctc_large—May 2020Conformer: Convolution-augmented Transformer for Speech … · code8.32
26Asr-conformer-loquacious—Feb 2025pwc-dump8.48
27Distil-small.en—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code8.57
28Whisper-small.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code8.59
29Distil-medium.en—Nov 2023Distil-Whisper: Robust Knowledge Distillation via Large-… · code8.77
30Niagara-38m-batch.en—Feb 2026pwc-dump8.91
31Stt_en_fastconformer_ctc_large—May 2023Fast Conformer with Linearly Scalable Attention for Effi…8.96
32Stt_en_fastconformer_transducer_large—May 2023Fast Conformer with Linearly Scalable Attention for Effi…9.06
33Moonshine-base—Oct 2024Moonshine: Speech Recognition for Live Transcription and… · code9.99
34Whisper-base.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code10.32
35Niagara-19m-batch.en—Feb 2026pwc-dump10.47
36Stt_en_conformer_ctc_small—May 2020Conformer: Convolution-augmented Transformer for Speech … · code11.16
37Moonshine-streaming-tiny—Jan 2026pwc-dump12
38Moonshine-tiny—Oct 2024Moonshine: Speech Recognition for Live Transcription and… · code12.65
39Whisper-tiny.en—Dec 2022Robust Speech Recognition via Large-Scale Weak Supervisi… · code12.81
40Asr-wav2vec2-librispeech—Jun 2021SpeechBrain: A General-Purpose Speech Toolkit · code14.35
41Wav2vec2-large-960h-lv60-self—Jun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of… · code21.27
42Mms-1b-all—May 2023Scaling Speech Technology to 1,000+ Languages · code22.54
43Hubert-xlarge-ls960-ft—Jun 2021HuBERT: Self-Supervised Speech Representation Learning b… · code22.55
44Hubert-large-ls960-ft—Jun 2021HuBERT: Self-Supervised Speech Representation Learning b… · code22.69
45Wav2vec2-large-robust-ft-libri-960h—Apr 2021Robust wav2vec 2.0: Analyzing Domain Shift in Self-Super… · code22.93
46Data2vec-audio-large-960h—Feb 2022data2vec: A General Framework for Self-supervised Learni… · code23.21
47Wav2vec2-conformer-rope-large-960h-ft—Oct 2020fairseq S2T: Fast Speech-to-Text Modeling with fairseq · code23.28
48Wav2vec2-conformer-rel-pos-large-960h-ft—Oct 2020fairseq S2T: Fast Speech-to-Text Modeling with fairseq · code23.29
49wav2vec 2.0 Large (960h)OpenMeta AIJun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of… · code26.77
50Data2vec-audio-base-960h—Feb 2022data2vec: A General Framework for Self-supervised Learni… · code28.30
51Wav2vec2-base-960h—Jun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of… · code29.40
52Mms-1b-fl102—May 2023Scaling Speech Technology to 1,000+ Languages · code39.80
Fig 2 · Rows sorted by score within each metric. Shaded row marks SOTA. Dates reflect model or paper release where available, otherwise the date Codesota accessed the source.
§ 03 · Progress

5 steps
of state of the art.

Each row below marks a model that broke the previous record on wer. Intermediate submissions are kept in the leaderboard above; only SOTA-setting entries are re-listed here.

Lower scores win. Each subsequent entry improved upon the previous best.

SOTA line · wer
  1. May 16, 2020Stt_en_conformer_ctc_large8.32
  2. Dec 6, 2022Whisper Large v3OpenAI7.44
  3. Apr 13, 2023Parakeet-tdt-0.6b-v26.05
  4. Mar 3, 2025Phi-4 Multimodal InstructMicrosoft6.02
  5. May 13, 2025Granite Speech 4.1 2BIBM5.33
Fig 3 · SOTA-setting models only. 5 entries span May 2020 → May 2025.
§ 04 · Literature

20 papers
tied to this benchmark.

Every paper below corresponds to at least one row in the leaderboard above. Click through for the arXiv preprint and, when available, the reference implementation.

§ 06 · Contribute

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

Submit a result ↵Read submission guide
What a submission needs
  • 01A public checkpoint or API endpoint
  • 02A reproduction script with frozen commit + seed
  • 03Declared evaluation environment (Python, deps)
  • 04One row per metric declared by this dataset
  • 05A contact so we can follow up on discrepancies