{"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/frill-a-non-semantic-speech-embedding-for","title":"FRILL: A Non-Semantic Speech Embedding for Mobile Devices","arxiv_id":"2011.04609","date":"2020-11-09","proceeding":null,"authors":["Jacob Peplinski","Joel Shor","Sachin Joglekar","Jake Garrison","Shwetak Patel"],"abstract":"Learned speech representations can drastically improve performance on tasks with limited labeled data. However, due to their size and complexity, learned representations have limited utility in mobile settings where run-time performance can be a significant bottleneck. In this work, we propose a class of lightweight non-semantic speech embedding models that run efficiently on mobile devices based on the recently proposed TRILL speech embedding. We combine novel architectural modifications with existing speed-up techniques to create embedding models that are fast enough to run in real-time on a mobile device and exhibit minimal performance degradation on a benchmark of non-semantic speech tasks. One such model (FRILL) is 32x faster on a Pixel 1 smartphone and 40% the size of TRILL, with an average decrease in accuracy of only 2%. To our knowledge, FRILL is the highest-quality non-semantic embedding designed for use on mobile devices. Furthermore, we demonstrate that these representations are useful for mobile health tasks such as non-speech human sounds detection and face-masked speech detection. Our models and code are publicly available.","url_abs":"https://arxiv.org/abs/2011.04609v5","url_pdf":"https://arxiv.org/pdf/2011.04609v5.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":"links_only","authors_date_abstract":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license), from the Kaggle arXiv metadata snapshot of 2026-09-12"},"code_links":[{"paper_slug":"frill-a-non-semantic-speech-embedding-for","repo_url":"https://github.com/google-research/google-research/tree/master/non_semantic_speech_benchmark/distillation","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[{"slug":"frill","name":"FRILL","full_name":"FRILL"}],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2011.04609","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}