{"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/optimization-of-speaker-extraction-neural","title":"Optimization of Speaker Extraction Neural Network with Magnitude and Temporal Spectrum Approximation Loss","arxiv_id":"1903.09952","date":"2019-03-24","proceeding":null,"authors":["Cheng-Lin Xu","Wei Rao","Eng Siong Chng","Haizhou Li"],"abstract":"The SpeakerBeam-FE (SBF) method is proposed for speaker extraction. It\nattempts to overcome the problem of unknown number of speakers in an audio\nrecording during source separation. The mask approximation loss of SBF is\nsub-optimal, which doesn't calculate direct signal reconstruction error and\nconsider the speech context. To address these problems, this paper proposes a\nmagnitude and temporal spectrum approximation loss to estimate a phase\nsensitive mask for the target speaker with the speaker characteristics.\nMoreover, this paper explores a concatenation framework instead of the context\nadaptive deep neural network in the SBF method to encode a speaker embedding\ninto the mask estimation network. Experimental results under open evaluation\ncondition show that the proposed method achieves 70.4% and 17.7% relative\nimprovement over the SBF baseline on signal-to-distortion ratio (SDR) and\nperceptual evaluation of speech quality (PESQ), respectively. A further\nanalysis demonstrates 69.1% and 72.3% relative SDR improvements obtained by the\nproposed method for different and same gender mixtures.","url_abs":"http://arxiv.org/abs/1903.09952v1","url_pdf":"http://arxiv.org/pdf/1903.09952v1.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":"optimization-of-speaker-extraction-neural","repo_url":"https://github.com/xuchenglin28/speaker_extraction","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}