{"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/noise-invariant-frame-selection-a-simple","title":"Noise Invariant Frame Selection: A Simple Method to Address the Background Noise Problem for Text-independent Speaker Verification","arxiv_id":"1805.01259","date":"2018-05-03","proceeding":null,"authors":["Siyang Song","Shuimei Zhang","Björn Schuller","Linlin Shen","Michel Valstar"],"abstract":"The performance of speaker-related systems usually degrades heavily in\npractical applications largely due to the presence of background noise. To\nimprove the robustness of such systems in unknown noisy environments, this\npaper proposes a simple pre-processing method called Noise Invariant Frame\nSelection (NIFS). Based on several noisy constraints, it selects noise\ninvariant frames from utterances to represent speakers. Experiments conducted\non the TIMIT database showed that the NIFS can significantly improve the\nperformance of Vector Quantization (VQ), Gaussian Mixture Model-Universal\nBackground Model (GMM-UBM) and i-vector-based speaker verification systems in\ndifferent unknown noisy environments with different SNRs, in comparison to\ntheir baselines. Meanwhile, the proposed NIFS-based speaker verification\nsystems achieves similar performance when we change the constraints\n(hyper-parameters) or features, which indicates that it is robust and easy to\nreproduce. Since NIFS is designed as a general algorithm, it could be further\napplied to other similar tasks.","url_abs":"http://arxiv.org/abs/1805.01259v1","url_pdf":"http://arxiv.org/pdf/1805.01259v1.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":"noise-invariant-frame-selection-a-simple","repo_url":"https://github.com/shuimove1234/Noise-Invariant-Frame-Selection","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"quantization","task_name":"Quantization"},{"task_slug":"speaker-verification","task_name":"Speaker Verification"},{"task_slug":"text-independent-speaker-verification","task_name":"Text-Independent Speaker Verification"}],"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}