Codesota · Benchmark · LibriSpeechHome/Leaderboards/Audio & Speech/Automatic Speech Recognition/LibriSpeech
Johns Hopkins University

LibriSpeech.

1000 hours of English speech from audiobooks. Standard benchmark for automatic speech recognition with clean and noisy test splits.

Paper ↗Lineage
§ 01 · Leaderboard

Results by metric.

No results yet on this benchmark
Help build the community leaderboard — submit your model results.
Found a wrong score or missing run?
Use row edits to send a sourced correction into moderation.
Add / edit result ↗Report issue ↗

No benchmark results available yet for LibriSpeech.

Check back soon as we continue collecting data.

Lineage

LibriSpeech in context.

See full speech recognition benchmarks lineage →
None — this is where the lineage begins.
This benchmark (1)
saturated2015-04
LibriSpeech
Successors (3)
active2017-06
VoxCeleb
VoxCeleb covers speaker identity, not transcription — a different task that addresses the 'who spoke' question LibriSpeech ignores. Speaker verification became a standard parallel track in speech evaluation.
active2020-04
CHiME-6
LibriSpeech test-other saturated; CHiME-6's multi-speaker dinner-party setup was the first major challenge where clean-speech progress didn't transfer. Where attention moved when LibriSpeech-other WER dropped below 4%.
active2021-06
GigaSpeech
GigaSpeech is a scale and diversity extension — 10× more data, multi-domain. A training and evaluation resource for robustness rather than a direct successor to LibriSpeech's narrow clean-speech task.
§ 04 · Submit a result

Add to the leaderboard.

Submit a Result

Sign in to submit benchmark results for LibriSpeech.

Sign in
← Back to Automatic Speech Recognition