{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,9,7]],"date-time":"2024-09-07T10:52:39Z","timestamp":1725706359431},"reference-count":51,"publisher":"IEEE","license":[{"start":{"date-parts":[[2023,12,16]],"date-time":"2023-12-16T00:00:00Z","timestamp":1702684800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,12,16]],"date-time":"2023-12-16T00:00:00Z","timestamp":1702684800000},"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":[[2023,12,16]]},"DOI":"10.1109\/asru57964.2023.10389619","type":"proceedings-article","created":{"date-parts":[[2024,1,19]],"date-time":"2024-01-19T18:38:40Z","timestamp":1705689520000},"page":"1-8","source":"Crossref","is-referenced-by-count":0,"title":["Audio-Visual Neural Syntax Acquisition"],"prefix":"10.1109","author":[{"given":"Cheng-I Jeff","family":"Lai","sequence":"first","affiliation":[{"name":"MIT"}]},{"given":"Freda","family":"Shi","sequence":"additional","affiliation":[{"name":"TTIC"}]},{"given":"Puyuan","family":"Peng","sequence":"additional","affiliation":[{"name":"UT Austin"}]},{"given":"Yoon","family":"Kim","sequence":"additional","affiliation":[{"name":"MIT"}]},{"given":"Kevin","family":"Gimpel","sequence":"additional","affiliation":[{"name":"TTIC"}]},{"given":"Shiyu","family":"Chang","sequence":"additional","affiliation":[{"name":"UC Santa Barbara"}]},{"given":"Yung-Sung","family":"Chuang","sequence":"additional","affiliation":[{"name":"MIT"}]},{"given":"Saurabhchand","family":"Bhati","sequence":"additional","affiliation":[{"name":"MIT"}]},{"given":"David","family":"Cox","sequence":"additional","affiliation":[{"name":"MIT-IBM Watson AI Lab"}]},{"given":"David","family":"Harwath","sequence":"additional","affiliation":[{"name":"UT Austin"}]},{"given":"Yang","family":"Zhang","sequence":"additional","affiliation":[{"name":"MIT-IBM Watson AI Lab"}]},{"given":"Karen","family":"Livescu","sequence":"additional","affiliation":[{"name":"TTIC"}]},{"given":"James","family":"Glass","sequence":"additional","affiliation":[{"name":"MIT"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cognition.2017.11.008"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2013.6639245"},{"article-title":"A generative constituent-context model for improved grammar induction","volume-title":"ACL","author":"Klein","key":"ref3","doi-asserted-by":"crossref","DOI":"10.3115\/1073083.1073106"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.3115\/1218955.1219016"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.4324\/9780203347089_chapter_5"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P17-1047"},{"journal-title":"ICLR","article-title":"Learning hierarchical discrete linguistic units from visually-grounded speech","year":"2020","author":"Harwath","key":"ref7"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1180"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10652"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1228"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-main.354"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.naacl-main.119"},{"article-title":"Unsupervised vision-language grammar induction with shared structure modeling","volume-title":"ICLR","author":"Wan","key":"ref13"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2007.909282"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.2011.6163965"},{"journal-title":"CogSci","article-title":"Unsupervised extraction of recurring words from infant-directed speech","year":"2011","author":"McInnes","key":"ref16"},{"volume-title":"Unsupervised speech processing with applications to query-by-example spoken term detection","year":"2013","author":"Zhang","key":"ref17"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00146"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1016\/j.csl.2017.04.008"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-1544"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-1874"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2022.3206084"},{"article-title":"Sparseval: Evaluation metrics for parsing speech","volume-title":"LREC","author":"Roark","key":"ref23"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1080\/01690961003589492"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-2530"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/N18-1007"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-3122"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-373"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/n19-1282"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P18-1249"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-381"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10094575"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/N19-1116"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2167"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1411"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/O-COCOSDA202152914.2021.9660456"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1007\/BF00992696"},{"journal-title":"arXiv preprint arXiv:1606.08415","article-title":"Gaussian error linear units (GELUs)","year":"2016","author":"Hendrycks","key":"ref38"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1016\/j.csl.2019.06.005"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-main.614"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1145\/6462.6502"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.acl-long.411"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/w16-3210"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-1386"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/taslp.2022.3229264"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3122291"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.3115\/112405.112467"},{"volume-title":"Structured tree alignment for evaluation of constituency parsing","year":"2023","author":"Shi","key":"ref49"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-2044"},{"volume-title":"The atoms of language","year":"2001","author":"Baker","key":"ref51"}],"event":{"name":"2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","start":{"date-parts":[[2023,12,16]]},"location":"Taipei, Taiwan","end":{"date-parts":[[2023,12,20]]}},"container-title":["2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10388490\/10389614\/10389619.pdf?arnumber=10389619","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,1,23]],"date-time":"2024-01-23T16:39:16Z","timestamp":1706027956000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10389619\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,12,16]]},"references-count":51,"URL":"https:\/\/doi.org\/10.1109\/asru57964.2023.10389619","relation":{},"subject":[],"published":{"date-parts":[[2023,12,16]]}}}
