Codesota · Speech · Speech Recognition · LibriSpeechTasks/Speech/Speech Recognition
Speech Recognition · benchmark dataset · 2015 · EN

LibriSpeech ASR Corpus.

1000 hours of English speech from audiobooks. Standard benchmark for automatic speech recognition.

Paper ↗Download datasetSubmit a result ↵
§ 01 · Leaderboard

Best published scores.

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

Primary
wer-test-clean · lower is better
All metrics
wer-test-clean, wer-test-other
wer-test-clean· primary
10 rows
#ModelOrgSubmittedPaper / codewer-test-clean
01Universal-1AssemblyAIApr 2024official1.60
02Conformer-CTC LargeOSSNVIDIA / NeMoJan 2023VALL-E: Neural Codec Language Models are Zero-Shot Text …1.70
03Parakeet-CTC-1.1BOSSNVIDIA / SunoNov 2023Parakeet: Efficient, Accurate Speech Recognition Adapted…1.70
04Canary-1BOSSNVIDIAOct 2023Canary: A Multilingual Speech Recognition Model1.70
05Whisper Large V3OSSOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi…1.80
06wav2vec 2.0 Large (960h)OSSMeta AIJun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of…1.80
07HuBERT Large (LS-960)OSSMeta AIJun 2021HuBERT: Self-Supervised Speech Representation Learning b…1.90
08Google USMGoogleMar 2023Google USM: Scaling Automatic Speech Recognition Beyond …2.00
09Whisper Large-v2OSSOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi…2.70
10Pulse STTSmallest AIMay 2026official3.22
wer-test-other
9 rows
#ModelOrgSubmittedPaper / codewer-test-other
01Universal-1AssemblyAIApr 2024official3.10
02wav2vec 2.0 Large (960h)OSSMeta AIJun 2020wav2vec 2.0: A Framework for Self-Supervised Learning of…3.30
03HuBERT Large (LS-960)OSSMeta AIJun 2021HuBERT: Self-Supervised Speech Representation Learning b…3.60
04Whisper Large V3OSSOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi…3.60
05Canary-1BOSSNVIDIAOct 2023Canary: A Multilingual Speech Recognition Model3.80
06Google USMGoogleMar 2023Google USM: Scaling Automatic Speech Recognition Beyond …4.10
07Parakeet-CTC-1.1BOSSNVIDIA / SunoNov 2023Parakeet: Efficient, Accurate Speech Recognition Adapted…4.20
08Whisper Large-v2OSSOpenAIDec 2022Robust Speech Recognition via Large-Scale Weak Supervisi…5.20
09Pulse STTSmallest AIMay 2026official5.83
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

3 steps
of state of the art.

Each row below marks a model that broke the previous record on wer-test-clean. 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-test-clean
  1. Jun 20, 2020wav2vec 2.0 Large (960h)Meta AI1.80
  2. Jan 5, 2023Conformer-CTC LargeNVIDIA / NeMo1.70
  3. Apr 3, 2024Universal-1AssemblyAI1.60
Fig 3 · SOTA-setting models only. 3 entries span Jun 2020 → Apr 2024.
§ 04 · Literature

7 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