{"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/filler-word-detection-and-classification-a","title":"Filler Word Detection and Classification: A Dataset and Benchmark","arxiv_id":"2203.15135","date":"2022-03-28","proceeding":null,"authors":["Ge Zhu","Juan-Pablo Caceres","Justin Salamon"],"abstract":"Filler words such as `uh' or `um' are sounds or words people use to signal they are pausing to think. Finding and removing filler words from recordings is a common and tedious task in media editing. Automatically detecting and classifying filler words could greatly aid in this task, but few studies have been published on this problem to date. A key reason is the absence of a dataset with annotated filler words for model training and evaluation. In this work, we present a novel speech dataset, PodcastFillers, with 35K annotated filler words and 50K annotations of other sounds that commonly occur in podcasts such as breaths, laughter, and word repetitions. We propose a pipeline that leverages VAD and ASR to detect filler candidates and a classifier to distinguish between filler word types. We evaluate our proposed pipeline on PodcastFillers, compare to several baselines, and present a detailed ablation study. In particular, we evaluate the importance of using ASR and how it compares to a transcription-free approach resembling keyword spotting. We show that our pipeline obtains state-of-the-art results, and that leveraging ASR strongly outperforms a keyword spotting approach. We make PodcastFillers publicly available, in the hope that our work serves as a benchmark for future research.","url_abs":"https://arxiv.org/abs/2203.15135v2","url_pdf":"https://arxiv.org/pdf/2203.15135v2.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":"filler-word-detection-and-classification-a","repo_url":"https://github.com/gzhu06/PodcastFillers_Utils","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"keyword-spotting","task_name":"Keyword Spotting"},{"task_slug":"sound-event-localization-and-detection","task_name":"Sound Event Localization and Detection"}],"methods":[],"datasets_introduced":[{"slug":"podcastfillers","name":"PodcastFillers","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/sound-event-localization-and-detection-on-2","task":"Sound Event Localization and Detection","dataset":"PodcastFillers","model":"AVC-FillerNet","rank_in_archive_order":1,"of":2,"metrics":{"event-based F1 score":"92.8"},"uses_additional_data":false},{"leaderboard":"/sota/sound-event-localization-and-detection-on-2","task":"Sound Event Localization and Detection","dataset":"PodcastFillers","model":"VC-FillerNet","rank_in_archive_order":2,"of":2,"metrics":{"event-based F1 score":"71.0"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}