{"about":{"site":"https://codewithpapers.app","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.","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"},"url":"/paper/semi-blind-source-separation-with","title":"Semi-blind source separation with multichannel variational autoencoder","arxiv_id":"1808.00892","date":"2018-08-02","proceeding":null,"authors":["Hirokazu Kameoka","Li Li","Shota Inoue","Shoji Makino"],"abstract":"This paper proposes a multichannel source separation technique called the\nmultichannel variational autoencoder (MVAE) method, which uses a conditional\nVAE (CVAE) to model and estimate the power spectrograms of the sources in a\nmixture. By training the CVAE using the spectrograms of training examples with\nsource-class labels, we can use the trained decoder distribution as a universal\ngenerative model capable of generating spectrograms conditioned on a specified\nclass label. By treating the latent space variables and the class label as the\nunknown parameters of this generative model, we can develop a\nconvergence-guaranteed semi-blind source separation algorithm that consists of\niteratively estimating the power spectrograms of the underlying sources as well\nas the separation matrices. In experimental evaluations, our MVAE produced\nbetter separation performance than a baseline method.","url_abs":"http://arxiv.org/abs/1808.00892v3","url_pdf":"http://arxiv.org/pdf/1808.00892v3.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":[{"paper_slug":"semi-blind-source-separation-with","repo_url":"https://github.com/mori97/MVAE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"blind-source-separation","task_name":"blind source separation"}],"methods":[{"method_slug":"cvae","method_name":"cVAE"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}