{"ID":2863382,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.24395","arxiv_id":"2509.24395","title":"Unsupervised Single-Channel Speech Separation with a Diffusion Prior under Speaker-Embedding Guidance","abstract":"Speech separation is a fundamental task in audio processing, typically addressed with fully supervised systems trained on paired mixtures. While effective, such systems typically rely on synthetic data pipelines, which may not reflect real-world conditions. Instead, we revisit the source-model paradigm, training a diffusion generative model solely on anechoic speech and formulating separation as a diffusion inverse problem. However, unconditional diffusion models lack speaker-level conditioning, they can capture local acoustic structure but produce temporally inconsistent speaker identities in separated sources. To address this limitation, we propose Speaker-Embedding guidance that, during the reverse diffusion process, maintains speaker coherence within each separated track while driving embeddings of different speakers further apart. In addition, we propose a new separation-oriented solver tailored for speech separation, and both strategies effectively enhance performance on the challenging task of unsupervised source-model-based speech separation, as confirmed by extensive experimental results. Audio samples and code are available at https://runwushi.github.io/UnSepDiff_demo.","short_abstract":"Speech separation is a fundamental task in audio processing, typically addressed with fully supervised systems trained on paired mixtures. While effective, such systems typically rely on synthetic data pipelines, which may not reflect real-world conditions. Instead, we revisit the source-model paradigm, training a diff...","url_abs":"https://arxiv.org/abs/2509.24395","url_pdf":"https://arxiv.org/pdf/2509.24395v1","authors":"[\"Runwu Shi\",\"Kai Li\",\"Chang Li\",\"Jiang Wang\",\"Sihan Tan\",\"Kazuhiro Nakadai\"]","published":"2025-09-29T07:42:54Z","proceeding":"eess.AS","tasks":"[\"eess.AS\",\"cs.SD\"]","methods":"[\"Diffusion Model\"]","has_code":false}