{"url":"/dataset/libri-light","name":"Libri-Light","full_name":null,"description_markdown":"Libri-Light is a collection of spoken English audio suitable for training speech recognition systems under limited or no supervision. It is derived from open-source audio books from the LibriVox project. It contains over 60K hours of audio.\r\n\r\nSource: [Libri-Light: A Benchmark for ASR with Limited or No Supervision](https://arxiv.org/pdf/1912.07875v1.pdf)","description_withheld":null,"homepage":"https://github.com/facebookresearch/libri-light","introduced_date":null,"introduced_date_note":null,"introduced_by":{"paper":"/paper/libri-light-a-benchmark-for-asr-with-limited","title":"Libri-Light: A Benchmark for ASR with Limited or No Supervision","first_author":"Jacob Kahn","url":null},"license":{"name":"Public domain","url":"https://wiki.librivox.org/index.php/Copyright_and_Public_Domain"},"modalities":[],"tasks":[{"name":"Speech Recognition","url":"/task/speech-recognition","datasets_with_task":"/datasets/task/speech-recognition"},{"name":"Language Modelling","url":"/task/language-modelling","datasets_with_task":"/datasets/task/language-modelling"}],"languages":[{"name":"English","url":"/datasets/language/english"}],"variants":["Libri-Light 60k, test-clean","Libri-Light 60k","Libri-Light test-clean","Libri-Light test-other","Libri-Light"],"data_loaders":[{"repo":"https://github.com/facebookresearch/libri-light","url":"https://github.com/facebookresearch/libri-light","frameworks":[]}],"num_papers_in_archive":194,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/speech-recognition-on-libri-light-test-clean","task":"Speech Recognition","dataset_variant":"Libri-Light test-clean","rows":5,"metrics":["Word Error Rate (WER)","ABX-across","ABX-within"],"first_row_in_archive_order":{"model":"wav2vec 2.0 Large-10h-LV-60k","paper":"/paper/wav2vec-2-0-a-framework-for-self-supervised","metrics":{"Word Error Rate (WER)":"2.5"},"code_links":[{"title":"huggingface/transformers","url":"https://github.com/huggingface/transformers"},{"title":"pytorch/fairseq","url":"https://github.com/pytorch/fairseq"},{"title":"pytorch/fairseq","url":"https://github.com/pytorch/fairseq/tree/master/examples/wav2vec"},{"title":"wenet-e2e/wenet","url":"https://github.com/wenet-e2e/wenet"},{"title":"sh-lee-prml/hierspeechpp","url":"https://github.com/sh-lee-prml/hierspeechpp"},{"title":"facebookresearch/brainmagick","url":"https://github.com/facebookresearch/brainmagick"},{"title":"mailong25/vietnamese-speech-recognition","url":"https://github.com/mailong25/vietnamese-speech-recognition"},{"title":"mailong25/self-supervised-speech-recognition","url":"https://github.com/mailong25/self-supervised-speech-recognition"},{"title":"huseinzol05/malaya-speech","url":"https://github.com/huseinzol05/malaya-speech"},{"title":"neonbjb/ocotillo","url":"https://github.com/neonbjb/ocotillo"},{"title":"shivangi-aneja/FaceTalk","url":"https://github.com/shivangi-aneja/FaceTalk"},{"title":"eastonYi/wav2vec","url":"https://github.com/eastonYi/wav2vec"},{"title":"vasudevgupta7/gsoc-wav2vec2","url":"https://github.com/vasudevgupta7/gsoc-wav2vec2"},{"title":"JoungheeKim/Non-Attentive-Tacotron","url":"https://github.com/JoungheeKim/Non-Attentive-Tacotron"},{"title":"HarunoriKawano/Wav2vec2.0","url":"https://github.com/HarunoriKawano/Wav2vec2.0"},{"title":"gatech-eic/s3-router","url":"https://github.com/gatech-eic/s3-router"},{"title":"BirgerMoell/tmh","url":"https://github.com/BirgerMoell/tmh"},{"title":"liutianlin0121/seislm","url":"https://github.com/liutianlin0121/seislm"},{"title":"AIdeaLab/wav2vec2_docker","url":"https://github.com/AIdeaLab/wav2vec2_docker"},{"title":"nlp-en-es/wav2vec2-spanish","url":"https://github.com/nlp-en-es/wav2vec2-spanish"},{"title":"phanxuanphucnd/wav2asr","url":"https://github.com/phanxuanphucnd/wav2asr"},{"title":"Arizona-Voice/Arizona-spotting","url":"https://github.com/Arizona-Voice/Arizona-spotting"},{"title":"phanxuanphucnd/Arizona-spotting","url":"https://github.com/phanxuanphucnd/Arizona-spotting"},{"title":"pwc-1/Paper-9","url":"https://github.com/pwc-1/Paper-9/tree/main/1/wav2vec2_conformer"},{"title":"phanxuanphucnd/Arizona-asr","url":"https://github.com/phanxuanphucnd/Arizona-asr"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/speech-recognition-on-libri-light-test-other","task":"Speech Recognition","dataset_variant":"Libri-Light test-other","rows":5,"metrics":["Word Error Rate (WER)","ABX-across","ABX-within"],"first_row_in_archive_order":{"model":"wav2vec 2.0 Large-10h-LV-60k","paper":"/paper/wav2vec-2-0-a-framework-for-self-supervised","metrics":{"Word Error Rate (WER)":"5.0"},"code_links":[{"title":"huggingface/transformers","url":"https://github.com/huggingface/transformers"},{"title":"pytorch/fairseq","url":"https://github.com/pytorch/fairseq"},{"title":"pytorch/fairseq","url":"https://github.com/pytorch/fairseq/tree/master/examples/wav2vec"},{"title":"wenet-e2e/wenet","url":"https://github.com/wenet-e2e/wenet"},{"title":"sh-lee-prml/hierspeechpp","url":"https://github.com/sh-lee-prml/hierspeechpp"},{"title":"facebookresearch/brainmagick","url":"https://github.com/facebookresearch/brainmagick"},{"title":"mailong25/vietnamese-speech-recognition","url":"https://github.com/mailong25/vietnamese-speech-recognition"},{"title":"mailong25/self-supervised-speech-recognition","url":"https://github.com/mailong25/self-supervised-speech-recognition"},{"title":"huseinzol05/malaya-speech","url":"https://github.com/huseinzol05/malaya-speech"},{"title":"neonbjb/ocotillo","url":"https://github.com/neonbjb/ocotillo"},{"title":"shivangi-aneja/FaceTalk","url":"https://github.com/shivangi-aneja/FaceTalk"},{"title":"eastonYi/wav2vec","url":"https://github.com/eastonYi/wav2vec"},{"title":"vasudevgupta7/gsoc-wav2vec2","url":"https://github.com/vasudevgupta7/gsoc-wav2vec2"},{"title":"JoungheeKim/Non-Attentive-Tacotron","url":"https://github.com/JoungheeKim/Non-Attentive-Tacotron"},{"title":"HarunoriKawano/Wav2vec2.0","url":"https://github.com/HarunoriKawano/Wav2vec2.0"},{"title":"gatech-eic/s3-router","url":"https://github.com/gatech-eic/s3-router"},{"title":"BirgerMoell/tmh","url":"https://github.com/BirgerMoell/tmh"},{"title":"liutianlin0121/seislm","url":"https://github.com/liutianlin0121/seislm"},{"title":"AIdeaLab/wav2vec2_docker","url":"https://github.com/AIdeaLab/wav2vec2_docker"},{"title":"nlp-en-es/wav2vec2-spanish","url":"https://github.com/nlp-en-es/wav2vec2-spanish"},{"title":"phanxuanphucnd/wav2asr","url":"https://github.com/phanxuanphucnd/wav2asr"},{"title":"Arizona-Voice/Arizona-spotting","url":"https://github.com/Arizona-Voice/Arizona-spotting"},{"title":"phanxuanphucnd/Arizona-spotting","url":"https://github.com/phanxuanphucnd/Arizona-spotting"},{"title":"pwc-1/Paper-9","url":"https://github.com/pwc-1/Paper-9/tree/main/1/wav2vec2_conformer"},{"title":"phanxuanphucnd/Arizona-asr","url":"https://github.com/phanxuanphucnd/Arizona-asr"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/wav2vec-2-0-a-framework-for-self-supervised","title":"wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations","date":"2020-06-20","rows_on_this_dataset":2,"code_links":25,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":9,"samples_ran":2,"samples_unverified":7,"pointer_only_for_licence":2,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/improving-unsupervised-sparsespeech-acoustic","title":"Improving Unsupervised Sparsespeech Acoustic Models with Categorical Reparameterization","date":"2020-05-29","rows_on_this_dataset":2,"code_links":1,"syntology":null},{"paper":"/paper/libri-light-a-benchmark-for-asr-with-limited","title":"Libri-Light: A Benchmark for ASR with Limited or No Supervision","date":"2019-12-17","rows_on_this_dataset":6,"code_links":2,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":14,"samples_ran":2,"samples_unverified":12,"pointer_only_for_licence":14,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":2,"samples_harvested":23,"samples_ran":4,"samples_unverified":19,"pointer_only_for_licence":16,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}