{"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/efficient-keyword-spotting-using-time-delay","title":"Efficient keyword spotting using time delay neural networks","arxiv_id":"1807.04353","date":"2018-08-28","proceeding":null,"authors":[],"abstract":"This paper describes a novel method of live keyword spotting using a\ntwo-stage time delay neural network. The model is trained using transfer\nlearning: initial training with phone targets from a large speech corpus is\nfollowed by training with keyword targets from a smaller data set. The accuracy\nof the system is evaluated on two separate tasks. The first is the freely\navailable Google Speech Commands dataset. The second is an in-house task\nspecifically developed for keyword spotting. The results show significant\nimprovements in false accept and false reject rates in both clean and noisy\nenvironments when compared with previously known techniques. Furthermore, we\ninvestigate various techniques to reduce computation in terms of\nmultiplications per second of audio. Compared to recently published work, the\nproposed system provides up to 89% savings on computational complexity.","url_abs":"http://arxiv.org/abs/1807.04353v2","url_pdf":"http://arxiv.org/pdf/1807.04353v2.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":"keyword-spotting","task_name":"Keyword Spotting"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/keyword-spotting-on-google-speech-commands","task":"Keyword Spotting","dataset":"Google Speech Commands","model":"TDNN","rank_in_archive_order":38,"of":42,"metrics":{"10-keyword Speech Commands dataset":"94.3"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}