{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T11:21:52Z","timestamp":1777634512590,"version":"3.51.4"},"reference-count":23,"publisher":"IEEE","license":[{"start":{"date-parts":[[2022,5,23]],"date-time":"2022-05-23T00:00:00Z","timestamp":1653264000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,5,23]],"date-time":"2022-05-23T00:00:00Z","timestamp":1653264000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,5,23]]},"DOI":"10.1109\/icassp43922.2022.9746684","type":"proceedings-article","created":{"date-parts":[[2022,4,27]],"date-time":"2022-04-27T19:50:34Z","timestamp":1651089034000},"page":"396-400","source":"Crossref","is-referenced-by-count":5,"title":["Musicyolo: A Sight-Singing Onset\/Offset Detection Framework Based on Object Detection Instead of Spectrum Frames"],"prefix":"10.1109","author":[{"given":"Xianke","family":"Wang","sequence":"first","affiliation":[{"name":"Huazhong University of Science and Technology,Hubei Provincial Key Laboratory of Smart Internet Technology,School of Electronic Information and Communications,Wuhan,China,430074"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wei","family":"Xu","sequence":"additional","affiliation":[{"name":"Huazhong University of Science and Technology,Hubei Provincial Key Laboratory of Smart Internet Technology,School of Electronic Information and Communications,Wuhan,China,430074"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Weiming","family":"Yang","sequence":"additional","affiliation":[{"name":"Huazhong University of Science and Technology,Hubei Provincial Key Laboratory of Smart Internet Technology,School of Electronic Information and Communications,Wuhan,China,430074"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wenqing","family":"Cheng","sequence":"additional","affiliation":[{"name":"Huazhong University of Science and Technology,Hubei Provincial Key Laboratory of Smart Internet Technology,School of Electronic Information and Communications,Wuhan,China,430074"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1749"},{"key":"ref11","first-page":"4160","article-title":"Sounddet: Polyphonic moving sound event detection and localization from raw waveform","author":"he","year":"2021","journal-title":"Int Conference on Machine Learning"},{"key":"ref12","article-title":"Yolox: Exceeding yolo series in 2021","author":"ge","year":"2021"},{"key":"ref13","first-page":"3","article-title":"Constant-q transform toolbox for music processing","author":"sch\u00f6rkhuber","year":"2010","journal-title":"Proceedings of Sound and Music Computing Conference"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.25080\/Majora-7b98e3ed-003"},{"key":"ref15","article-title":"labelme: Image polygonal annotation with python","author":"kentaro","year":"2016"},{"key":"ref16","article-title":"Yolov3: An incremental improvement","author":"redmon","year":"2018"},{"key":"ref17","first-page":"567","article-title":"Evaluation framework for automatic singing transcription","author":"molina","year":"2014","journal-title":"Proceedings of the International Society for Music Information Retrieval Conference (ISMIR)"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2014.2331102"},{"key":"ref19","article-title":"Computeraided melody note transcription using the tony software: Accuracy and efficiency","author":"mauch","year":"2015","journal-title":"Proceedings of the First International Conference on Technologies for Music Notation and Representation"},{"key":"ref4","article-title":"Power-scaled spectral flux and peak-valley group-delay methods for robust musical onset detection","author":"su","year":"2014","journal-title":"ICMC"},{"key":"ref3","first-page":"361","article-title":"Local group delay based vibrato and tremolo suppression for onset detection","author":"b\u00f6ck","year":"2013","journal-title":"Proceedings of the International Society for Music Information Retrieval Conference (ISMIR)"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.33682\/4jhy-bj81"},{"key":"ref5","article-title":"Enabling factorized piano music modeling and generation with the maestro dataset","author":"hawthorne","year":"2019","journal-title":"Proceedings of International Conference on Learning Representations"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-3097"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462062"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2014.6853672"},{"key":"ref1","first-page":"560","article-title":"Onset detection in composition items of carnatic music","author":"sebastian","year":"2017","journal-title":"IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP)"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682376"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7952166"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414601"},{"key":"ref21","first-page":"367","article-title":"mir eval: A transparent implementation of common mir metrics","author":"raffel","year":"2014","journal-title":"Proceedings of the International Society for Music Information Retrieval Conference (ISMIR)"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2010.2042119"}],"event":{"name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Singapore, Singapore","start":{"date-parts":[[2022,5,23]]},"end":{"date-parts":[[2022,5,27]]}},"container-title":["ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9745891\/9746004\/09746684.pdf?arnumber=9746684","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,15]],"date-time":"2022-08-15T20:08:19Z","timestamp":1660594099000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9746684\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,5,23]]},"references-count":23,"URL":"https:\/\/doi.org\/10.1109\/icassp43922.2022.9746684","relation":{},"subject":[],"published":{"date-parts":[[2022,5,23]]}}}