{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,27]],"date-time":"2025-10-27T16:22:44Z","timestamp":1761582164735},"reference-count":28,"publisher":"IEEE","license":[{"start":{"date-parts":[[2021,12,13]],"date-time":"2021-12-13T00:00:00Z","timestamp":1639353600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2021,12,13]],"date-time":"2021-12-13T00:00:00Z","timestamp":1639353600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2021,12,13]]},"DOI":"10.1109\/asru51503.2021.9687967","type":"proceedings-article","created":{"date-parts":[[2022,2,3]],"date-time":"2022-02-03T20:31:00Z","timestamp":1643920260000},"page":"571-578","source":"Crossref","is-referenced-by-count":8,"title":["Multi-Task Learning with Cross Attention for Keyword Spotting"],"prefix":"10.1109","author":[{"given":"Takuya","family":"Higuchil","sequence":"first","affiliation":[{"name":"Apple"}]},{"given":"Anmol","family":"Gupta","sequence":"additional","affiliation":[{"name":"The University of Hong Kong,Department of Computer Science"}]},{"given":"Chandra","family":"Dhir","sequence":"additional","affiliation":[{"name":"Apple"}]}],"member":"263","reference":[{"key":"ref10","first-page":"220","article-title":"An application of recurrent neural networks to discriminative keyword spotting","author":"fern\u00e1ndez","year":"2007","journal-title":"International Conference on Artificial Neural Networks"},{"key":"ref11","first-page":"1606","article-title":"Convolutional recurrent neural networks for small-footprint keyword spotting","author":"arik","year":"0","journal-title":"InterSpeech"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-1393"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8683557"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-2763"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1230"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-753"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7953164"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.2017.8268974"},{"key":"ref19","first-page":"3351","article-title":"Hybrid transformer\/etc net-works for hardware efficient voice triggering","author":"adya","year":"0","journal-title":"InterSpeech"},{"key":"ref28","article-title":"Adam: A method for stochastic optimization","author":"kingma","year":"2014","journal-title":"ArXiv Preprint"},{"key":"ref4","first-page":"3607","article-title":"Compressed time delay neural network for small-footprint keyword spotting","author":"ming","year":"0","journal-title":"InterSpeech"},{"key":"ref27","article-title":"A cascade architecture for keyword spotting on mobile devices","author":"gruenstein","year":"2017","journal-title":"ArXiv Preprint"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-1485"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462166"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.2017.8268943"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-2204"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462227"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178863"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2015-352"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2014.6854370"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053577"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1186"},{"key":"ref21","article-title":"Progressive voice trigger detection: Accuracy vs latency","author":"sigtia","year":"2020","journal-title":"ArXiv Preprint"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU46091.2019.9003750"},{"key":"ref23","first-page":"1","article-title":"Attention is all you need","author":"ashish","year":"0","journal-title":"31st Conference on Neural Information Processing Systems (NIPS 2017)"},{"key":"ref26","article-title":"Layer normalization","author":"ba","year":"2016","journal-title":"ArXiv Preprint"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414339"}],"event":{"name":"2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","start":{"date-parts":[[2021,12,13]]},"location":"Cartagena, Colombia","end":{"date-parts":[[2021,12,17]]}},"container-title":["2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9687821\/9687855\/09687967.pdf?arnumber=9687967","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,5,16]],"date-time":"2022-05-16T20:41:40Z","timestamp":1652733700000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9687967\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,12,13]]},"references-count":28,"URL":"https:\/\/doi.org\/10.1109\/asru51503.2021.9687967","relation":{},"subject":[],"published":{"date-parts":[[2021,12,13]]}}}