{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,24]],"date-time":"2026-06-24T15:00:42Z","timestamp":1782313242524,"version":"3.54.5"},"reference-count":48,"publisher":"IEEE","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2017,12]]},"DOI":"10.1109\/asru.2017.8268974","type":"proceedings-article","created":{"date-parts":[[2018,1,25]],"date-time":"2018-01-25T21:43:53Z","timestamp":1516916633000},"page":"474-481","source":"Crossref","is-referenced-by-count":45,"title":["Streaming small-footprint keyword spotting using sequence-to-sequence models"],"prefix":"10.1109","author":[{"given":"Yanzhang","family":"He","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rohit","family":"Prabhavalkar","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kanishka","family":"Rao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anton","family":"Bakhtin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ian","family":"McGraw","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178962"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/APSIPA.2013.6694175"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472820"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7953076"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7953164"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1990.115555"},{"key":"ref36","first-page":"302","author":"sz\u00f6ke","year":"2005","journal-title":"Phoneme Based Acoustics Keyword Spotting in Informal Continuous Speech"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.2015.7404790"},{"key":"ref34","article-title":"Fast and accurate recurrent neural network acoustic models for speech recognition","author":"sak","year":"2015","journal-title":"Proc of Interspeech"},{"key":"ref10","first-page":"220","article-title":"An application of recurrent neural networks to discriminative keyword spotting","author":"fern\u00e1ndez","year":"2007","journal-title":"Proc of International Conference on Artificial Neural Networks (ICANN)"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1993.319341"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-234"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1016\/j.specom.2008.10.002"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2013.6639281"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.2009.5372889"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178970"},{"key":"ref16","article-title":"Sequence transduction with recurrent neural networks","author":"graves","year":"2012","journal-title":"Proc of ICASSP"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2013.6638947"},{"key":"ref18","doi-asserted-by":"crossref","DOI":"10.21437\/Interspeech.2017-1705","article-title":"Recurrent neural aligner: An encoder-decoder neural network model for sequence-to-sequence mapping","author":"sak","year":"2017","journal-title":"Proc of Interspeech"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/1143844.1143891"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-232"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2014.6854370"},{"key":"ref27","article-title":"A comparison of sequence-to-sequence models for speech recognition","author":"prabhavalkar","year":"2017","journal-title":"Proc of Interspeech"},{"key":"ref3","article-title":"The SRI\/OGI 2006 spoken term detection system","author":"vergyri","year":"2007","journal-title":"Proc of Interspeech"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-1393"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-753"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178863"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/SLT.2016.7846306"},{"key":"ref7","article-title":"Convolutional neural networks for small footprint keyword spotting","author":"sainath","year":"2015","journal-title":"Proc of Interspeech"},{"key":"ref2","article-title":"Rapid and accurate spoken term detection","author":"miller","year":"2007","journal-title":"Proc of Interspeech"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-1737"},{"key":"ref1","first-page":"51","article-title":"results of the 2006 spoken term detection evaluation","author":"fiscus","year":"2007","journal-title":"Proc Special Interest Group on Information Retrieval (SIGIR)"},{"key":"ref46","article-title":"Lstm: A search space odyssey","author":"greff","year":"2016","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"key":"ref20","article-title":"Online keyword spotting with a character-level recurrent neural network","author":"hwang","year":"2015","journal-title":"ArXiv e-prints"},{"key":"ref45","doi-asserted-by":"crossref","DOI":"10.21437\/Interspeech.2017-639","article-title":"Optimizing expected word error rate via sampling for speech recognition","author":"shannon","year":"2017","journal-title":"Proc of Interspeech"},{"key":"ref48","author":"abadi","year":"2015","journal-title":"Tensorflow Large-scale machine learning on heterogeneous distributed systems"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ISCSLP.2016.7918460"},{"key":"ref47","first-page":"1223","article-title":"Large Scale Distributed Deep Networks","author":"dean","year":"2012","journal-title":"Proc of NIPS"},{"key":"ref21","article-title":"An end-to-end architecture for keyword spotting and voice activity detection","author":"lengerich","year":"2016","journal-title":"ArXiv e-prints"},{"key":"ref42","doi-asserted-by":"crossref","DOI":"10.21437\/Interspeech.2012-498","article-title":"White listing and score normalization for keyword spotting of noisy speech","author":"zhang","year":"2012","journal-title":"Proc of Interspeech"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472621"},{"key":"ref41","article-title":"Confidence intervals for the area under the roc curve","author":"cortes","year":"2004","journal-title":"Proc of NIPS"},{"key":"ref23","article-title":"Neural speech recognizer: Acoustic-to-word lstm model for large vocabulary speech recognition","author":"soltau","year":"2016","journal-title":"ArXiv e-prints"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472823"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472641"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-128"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472618"}],"event":{"name":"2017 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","location":"Okinawa, Japan","start":{"date-parts":[[2017,12,16]]},"end":{"date-parts":[[2017,12,20]]}},"container-title":["2017 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/8260578\/8268903\/08268974.pdf?arnumber=8268974","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,13]],"date-time":"2022-08-13T06:17:38Z","timestamp":1660371458000},"score":1,"resource":{"primary":{"URL":"http:\/\/ieeexplore.ieee.org\/document\/8268974\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2017,12]]},"references-count":48,"URL":"https:\/\/doi.org\/10.1109\/asru.2017.8268974","relation":{},"subject":[],"published":{"date-parts":[[2017,12]]}}}