Jointly Encoding Word Confusion Network and Dialogue Context with BERT for Spoken Language Understanding

Spoken Language Understanding (SLU) converts hypotheses from automatic speech recognizer (ASR) into structured semantic representations. ASR recognition errors can severely degenerate the performance of the subsequent SLU module. To address this issue, word confusion networks (WCNs) have been used t...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2020-09
Main Authors: Liu, Chen, Zhu, Su, Zhao, Zijian, Cao, Ruisheng, Chen, Lu, Yu, Kai
Format: Article
Language:English
Subjects:
Confusion
Context
Hypotheses
Speech recognition
Words (language)
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Liu, Chen
Zhu, Su
Zhao, Zijian
Cao, Ruisheng
Chen, Lu
Yu, Kai
description Spoken Language Understanding (SLU) converts hypotheses from automatic speech recognizer (ASR) into structured semantic representations. ASR recognition errors can severely degenerate the performance of the subsequent SLU module. To address this issue, word confusion networks (WCNs) have been used to encode the input for SLU, which contain richer information than 1-best or n-best hypotheses list. To further eliminate ambiguity, the last system act of dialogue context is also utilized as additional input. In this paper, a novel BERT based SLU model (WCN-BERT SLU) is proposed to encode WCNs and the dialogue context jointly. It can integrate both structural information and ASR posterior probabilities of WCNs in the BERT architecture. Experiments on DSTC2, a benchmark of SLU, show that the proposed method is effective and can outperform previous state-of-the-art models significantly.
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2406738495</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2406738495</sourcerecordid><originalsourceid>FETCH-proquest_journals_24067384953</originalsourceid><addsrcrecordid>eNqNjMsKwjAQRYMgWNR_GHBdqEnrY6tWRMSFVlxKsGlNLTM1D9S_t4If4Ooszrm3wwIuxDicxZz32NDaKooiPpnyJBEBu21Jo6vfkOKVco0lnMnksCQsvNWEsFfuSeYOEnNYaVlT6dVXO_Vy8NTuBov0kEFBBo4N3RXCTmLpZanghLky1rXL9nfAuoWsrRr-2GejdZotN2Fj6OGVdZeKvMFWXXgcTaZiFs8T8V_1ASauR5M</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2406738495</pqid></control><display><type>article</type><title>Jointly Encoding Word Confusion Network and Dialogue Context with BERT for Spoken Language Understanding</title><source>Publicly Available Content Database</source><creator>Liu, Chen ; Zhu, Su ; Zhao, Zijian ; Cao, Ruisheng ; Chen, Lu ; Yu, Kai</creator><creatorcontrib>Liu, Chen ; Zhu, Su ; Zhao, Zijian ; Cao, Ruisheng ; Chen, Lu ; Yu, Kai</creatorcontrib><description>Spoken Language Understanding (SLU) converts hypotheses from automatic speech recognizer (ASR) into structured semantic representations. ASR recognition errors can severely degenerate the performance of the subsequent SLU module. To address this issue, word confusion networks (WCNs) have been used to encode the input for SLU, which contain richer information than 1-best or n-best hypotheses list. To further eliminate ambiguity, the last system act of dialogue context is also utilized as additional input. In this paper, a novel BERT based SLU model (WCN-BERT SLU) is proposed to encode WCNs and the dialogue context jointly. It can integrate both structural information and ASR posterior probabilities of WCNs in the BERT architecture. Experiments on DSTC2, a benchmark of SLU, show that the proposed method is effective and can outperform previous state-of-the-art models significantly.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Confusion ; Context ; Hypotheses ; Speech recognition ; Words (language)</subject><ispartof>arXiv.org, 2020-09</ispartof><rights>2020. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2406738495?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>776,780,25731,36989,44566</link.rule.ids></links><search><creatorcontrib>Liu, Chen</creatorcontrib><creatorcontrib>Zhu, Su</creatorcontrib><creatorcontrib>Zhao, Zijian</creatorcontrib><creatorcontrib>Cao, Ruisheng</creatorcontrib><creatorcontrib>Chen, Lu</creatorcontrib><creatorcontrib>Yu, Kai</creatorcontrib><title>Jointly Encoding Word Confusion Network and Dialogue Context with BERT for Spoken Language Understanding</title><title>arXiv.org</title><description>Spoken Language Understanding (SLU) converts hypotheses from automatic speech recognizer (ASR) into structured semantic representations. ASR recognition errors can severely degenerate the performance of the subsequent SLU module. To address this issue, word confusion networks (WCNs) have been used to encode the input for SLU, which contain richer information than 1-best or n-best hypotheses list. To further eliminate ambiguity, the last system act of dialogue context is also utilized as additional input. In this paper, a novel BERT based SLU model (WCN-BERT SLU) is proposed to encode WCNs and the dialogue context jointly. It can integrate both structural information and ASR posterior probabilities of WCNs in the BERT architecture. Experiments on DSTC2, a benchmark of SLU, show that the proposed method is effective and can outperform previous state-of-the-art models significantly.</description><subject>Confusion</subject><subject>Context</subject><subject>Hypotheses</subject><subject>Speech recognition</subject><subject>Words (language)</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNqNjMsKwjAQRYMgWNR_GHBdqEnrY6tWRMSFVlxKsGlNLTM1D9S_t4If4Ooszrm3wwIuxDicxZz32NDaKooiPpnyJBEBu21Jo6vfkOKVco0lnMnksCQsvNWEsFfuSeYOEnNYaVlT6dVXO_Vy8NTuBov0kEFBBo4N3RXCTmLpZanghLky1rXL9nfAuoWsrRr-2GejdZotN2Fj6OGVdZeKvMFWXXgcTaZiFs8T8V_1ASauR5M</recordid><startdate>20200908</startdate><enddate>20200908</enddate><creator>Liu, Chen</creator><creator>Zhu, Su</creator><creator>Zhao, Zijian</creator><creator>Cao, Ruisheng</creator><creator>Chen, Lu</creator><creator>Yu, Kai</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20200908</creationdate><title>Jointly Encoding Word Confusion Network and Dialogue Context with BERT for Spoken Language Understanding</title><author>Liu, Chen ; Zhu, Su ; Zhao, Zijian ; Cao, Ruisheng ; Chen, Lu ; Yu, Kai</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_24067384953</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Confusion</topic><topic>Context</topic><topic>Hypotheses</topic><topic>Speech recognition</topic><topic>Words (language)</topic><toplevel>online_resources</toplevel><creatorcontrib>Liu, Chen</creatorcontrib><creatorcontrib>Zhu, Su</creatorcontrib><creatorcontrib>Zhao, Zijian</creatorcontrib><creatorcontrib>Cao, Ruisheng</creatorcontrib><creatorcontrib>Chen, Lu</creatorcontrib><creatorcontrib>Yu, Kai</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Liu, Chen</au><au>Zhu, Su</au><au>Zhao, Zijian</au><au>Cao, Ruisheng</au><au>Chen, Lu</au><au>Yu, Kai</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Jointly Encoding Word Confusion Network and Dialogue Context with BERT for Spoken Language Understanding</atitle><jtitle>arXiv.org</jtitle><date>2020-09-08</date><risdate>2020</risdate><eissn>2331-8422</eissn><abstract>Spoken Language Understanding (SLU) converts hypotheses from automatic speech recognizer (ASR) into structured semantic representations. ASR recognition errors can severely degenerate the performance of the subsequent SLU module. To address this issue, word confusion networks (WCNs) have been used to encode the input for SLU, which contain richer information than 1-best or n-best hypotheses list. To further eliminate ambiguity, the last system act of dialogue context is also utilized as additional input. In this paper, a novel BERT based SLU model (WCN-BERT SLU) is proposed to encode WCNs and the dialogue context jointly. It can integrate both structural information and ASR posterior probabilities of WCNs in the BERT architecture. Experiments on DSTC2, a benchmark of SLU, show that the proposed method is effective and can outperform previous state-of-the-art models significantly.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2020-09
issn 2331-8422
language eng
recordid cdi_proquest_journals_2406738495
source Publicly Available Content Database
subjects Confusion
Context
Hypotheses
Speech recognition
Words (language)
title Jointly Encoding Word Confusion Network and Dialogue Context with BERT for Spoken Language Understanding
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-30T19%3A18%3A58IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Jointly%20Encoding%20Word%20Confusion%20Network%20and%20Dialogue%20Context%20with%20BERT%20for%20Spoken%20Language%20Understanding&rft.jtitle=arXiv.org&rft.au=Liu,%20Chen&rft.date=2020-09-08&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2406738495%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_24067384953%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2406738495&rft_id=info:pmid/&rfr_iscdi=true