{"about":{"non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","site":"https://codewithpapers.app","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/separate-and-diffuse-using-a-pretrained","title":"Separate And Diffuse: Using a Pretrained Diffusion Model for Improving Source Separation","arxiv_id":"2301.10752","date":"2023-01-25","proceeding":null,"authors":["Shahar Lutati","Eliya Nachmani","Lior Wolf"],"abstract":"The problem of speech separation, also known as the cocktail party problem, refers to the task of isolating a single speech signal from a mixture of speech signals. Previous work on source separation derived an upper bound for the source separation task in the domain of human speech. This bound is derived for deterministic models. Recent advancements in generative models challenge this bound. We show how the upper bound can be generalized to the case of random generative models. Applying a diffusion model Vocoder that was pretrained to model single-speaker voices on the output of a deterministic separation model leads to state-of-the-art separation results. It is shown that this requires one to combine the output of the separation model with that of the diffusion model. In our method, a linear combination is performed, in the frequency domain, using weights that are inferred by a learned model. We show state-of-the-art results on 2, 3, 5, 10, and 20 speakers on multiple benchmarks. In particular, for two speakers, our method is able to surpass what was previously considered the upper performance bound.","url_abs":"https://arxiv.org/abs/2301.10752v2","url_pdf":"https://arxiv.org/pdf/2301.10752v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"audio-source-separation","task_name":"Audio Source Separation"},{"task_slug":"generalization-bounds","task_name":"Generalization Bounds"},{"task_slug":"multi-speaker-source-separation","task_name":"Multi-Speaker Source Separation"},{"task_slug":"speech-separation","task_name":"Speech Separation"}],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"diffusion","method_name":"Diffusion"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"prelu","method_name":"PReLU"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"sepformer","method_name":"SepFormer"},{"method_slug":"separate-and-diffuse","method_name":"Separate And Diffuse"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[{"slug":"separate-and-diffuse","name":"Separate And Diffuse","full_name":"Separate And Diffuse"}],"results":[{"leaderboard":"/sota/speech-separation-on-libri10mix","task":"Speech Separation","dataset":"Libri10Mix","model":"Separate And Diffuse","rank_in_archive_order":1,"of":3,"metrics":{"SI-SDRi":"9"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-libri20mix","task":"Speech Separation","dataset":"Libri20Mix","model":"Separate And Diffuse","rank_in_archive_order":1,"of":2,"metrics":{"SI-SDRi":"5.2"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-libri2mix","task":"Speech Separation","dataset":"Libri2Mix","model":"Separate And Diffuse","rank_in_archive_order":4,"of":10,"metrics":{"SI-SDRi":"21.5"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-libri5mix","task":"Speech Separation","dataset":"Libri5Mix","model":"Separate And Diffuse","rank_in_archive_order":1,"of":4,"metrics":{"SI-SDRi":"14.2"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-wsj0-2mix","task":"Speech Separation","dataset":"WSJ0-2mix","model":"Separate And Diffuse","rank_in_archive_order":7,"of":40,"metrics":{"SI-SDRi":"23.9"},"uses_additional_data":false},{"leaderboard":"/sota/speech-separation-on-wsj0-3mix","task":"Speech Separation","dataset":"WSJ0-3mix","model":"Separate And Diffuse","rank_in_archive_order":4,"of":9,"metrics":{"SI-SDRi":"20.9"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2301.10752","atlas_url":"https://app.syntology.ai/?focus=2301.10752","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"2301.10752"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2301.10752\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2301.10752"},"call":"get_concepts_for_paper(arxiv_id=\"2301.10752\")"}],"by_repo":[],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":"the archive holds no code link for this paper","arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"276be2a3f21bba51844821d4b7f444bf1395fe5a40f37effe55e58ac152aded0","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}