{"url":"/dataset/librimix","name":"LibriMix","full_name":null,"description_markdown":"LibriMix is an open-source alternative to wsj0-2mix. Based on LibriSpeech, LibriMix consists of two- or three-speaker mixtures combined with ambient noise samples from WHAM!. \r\n\r\nSource: [LibriMix: An Open-Source Dataset for Generalizable Speech Separation](/paper/librimix-an-open-source-dataset-for)","description_withheld":null,"homepage":"https://github.com/JorisCos/LibriMix","introduced_date":null,"introduced_date_note":null,"introduced_by":{"paper":"/paper/librimix-an-open-source-dataset-for","title":"LibriMix: An Open-Source Dataset for Generalizable Speech Separation","first_author":null,"url":null},"license":null,"modalities":[{"name":"Speech","url":"/datasets/modality/speech"}],"tasks":[{"name":"Speech Enhancement","url":"/task/speech-enhancement","datasets_with_task":"/datasets/task/speech-enhancement"},{"name":"Speech Separation","url":"/task/speech-separation","datasets_with_task":"/datasets/task/speech-separation"},{"name":"Audio Source Separation","url":"/task/audio-source-separation","datasets_with_task":"/datasets/task/audio-source-separation"}],"languages":[],"variants":["LibriMix","Libri2Mix"],"data_loaders":[{"repo":"https://github.com/JorisCos/LibriMix","url":"https://github.com/JorisCos/LibriMix","frameworks":[]}],"num_papers_in_archive":122,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/speech-separation-on-libri2mix","task":"Speech Separation","dataset_variant":"Libri2Mix","rows":10,"metrics":["SI-SDRi","SDRi","Number of parameters (M)","SDR"],"first_row_in_archive_order":{"model":"MossFormer2 (w speed perturb)","paper":"/paper/mossformer2-combining-transformer-and-rnn-1","metrics":{"SI-SDRi":"22.2"},"code_links":[{"title":"modelscope/ClearerVoice-Studio","url":"https://github.com/modelscope/ClearerVoice-Studio"},{"title":"alibabasglab/MossFormer2","url":"https://github.com/alibabasglab/MossFormer2"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/wanna-hear-your-voice-adaptive-effective-and","title":"Wanna hear your voice? A sample is all we need!","date":"2024-10-01","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/tf-locoformer-transformer-with-local-modeling","title":"TF-Locoformer: Transformer with Local Modeling by Convolution for Speech Separation and Enhancement","date":"2024-08-06","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/separate-and-diffuse-using-a-pretrained","title":"Separate And Diffuse: Using a Pretrained Diffusion Model for Improving Source Separation","date":"2023-01-25","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/an-efficient-encoder-decoder-architecture","title":"An efficient encoder-decoder architecture with top-down attention for speech separation","date":"2022-09-30","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":18,"samples_ran":13,"samples_unverified":5,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/self-supervised-pre-training-reduces-label","title":"Stabilizing Label Assignment for Speech Separation by Self-supervised Pre-training","date":"2020-10-29","rows_on_this_dataset":3,"code_links":1,"syntology":null},{"paper":"/paper/mossformer2-combining-transformer-and-rnn-1","title":"MossFormer2: Combining Transformer and RNN-Free Recurrent Network for Enhanced Time-Domain Monaural Speech Separation","date":null,"rows_on_this_dataset":2,"code_links":2,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":4,"samples_ran":4,"samples_unverified":0,"pointer_only_for_licence":0,"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":22,"samples_ran":17,"samples_unverified":5,"pointer_only_for_licence":0,"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."}