Enabling accent resilient speech based information retrieval

Koushik Sinha, Geetha Manjunath, Raveesh R. Sharma, Viswanath Gangavaram, A. Pooja, Deepak R. Murugaian

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Voice interfaces to browsers and mobile applications are becoming popular as typing with touch screens is cumbersome. The main issue of practical speech based interfaces is how to overcome speech recognition errors. This problem is more severe when the users are non-native speakers of English due to differences in pronunciations. In this paper, we describe a novel, intelligent speech interface design approach for IR tasks that is significantly robust to accent variations. Our solution uses phonemic similarity based word spreading and semantic information based filtering to boost the accuracy of any ASR. We evaluated our solution with Google Voice as the ASR for a web question-answering system developed in-house and the results are very encouraging. Copyright is held by the author/owner(s).

Original languageEnglish
Title of host publicationWWW'12 - Proceedings of the 21st Annual Conference on World Wide Web Companion
Pages601-602
Number of pages2
DOIs
Publication statusPublished - 21 May 2012
Externally publishedYes
Event21st Annual Conference on World Wide Web, WWW'12 - Lyon, France
Duration: 16 Apr 201220 Apr 2012

Other

Other21st Annual Conference on World Wide Web, WWW'12
CountryFrance
CityLyon
Period16/4/1220/4/12

Fingerprint

Information retrieval
Touch screens
Speech recognition
Semantics

Keywords

  • Phonemic spreading
  • QA
  • Semantic feedback
  • Speech based IR

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Sinha, K., Manjunath, G., Sharma, R. R., Gangavaram, V., Pooja, A., & Murugaian, D. R. (2012). Enabling accent resilient speech based information retrieval. In WWW'12 - Proceedings of the 21st Annual Conference on World Wide Web Companion (pp. 601-602) https://doi.org/10.1145/2187980.2188147

Enabling accent resilient speech based information retrieval. / Sinha, Koushik; Manjunath, Geetha; Sharma, Raveesh R.; Gangavaram, Viswanath; Pooja, A.; Murugaian, Deepak R.

WWW'12 - Proceedings of the 21st Annual Conference on World Wide Web Companion. 2012. p. 601-602.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Sinha, K, Manjunath, G, Sharma, RR, Gangavaram, V, Pooja, A & Murugaian, DR 2012, Enabling accent resilient speech based information retrieval. in WWW'12 - Proceedings of the 21st Annual Conference on World Wide Web Companion. pp. 601-602, 21st Annual Conference on World Wide Web, WWW'12, Lyon, France, 16/4/12. https://doi.org/10.1145/2187980.2188147
Sinha K, Manjunath G, Sharma RR, Gangavaram V, Pooja A, Murugaian DR. Enabling accent resilient speech based information retrieval. In WWW'12 - Proceedings of the 21st Annual Conference on World Wide Web Companion. 2012. p. 601-602 https://doi.org/10.1145/2187980.2188147
Sinha, Koushik ; Manjunath, Geetha ; Sharma, Raveesh R. ; Gangavaram, Viswanath ; Pooja, A. ; Murugaian, Deepak R. / Enabling accent resilient speech based information retrieval. WWW'12 - Proceedings of the 21st Annual Conference on World Wide Web Companion. 2012. pp. 601-602
@inproceedings{cff9cb0288e048109e783fdbdb663c3b,
title = "Enabling accent resilient speech based information retrieval",
abstract = "Voice interfaces to browsers and mobile applications are becoming popular as typing with touch screens is cumbersome. The main issue of practical speech based interfaces is how to overcome speech recognition errors. This problem is more severe when the users are non-native speakers of English due to differences in pronunciations. In this paper, we describe a novel, intelligent speech interface design approach for IR tasks that is significantly robust to accent variations. Our solution uses phonemic similarity based word spreading and semantic information based filtering to boost the accuracy of any ASR. We evaluated our solution with Google Voice as the ASR for a web question-answering system developed in-house and the results are very encouraging. Copyright is held by the author/owner(s).",
keywords = "Phonemic spreading, QA, Semantic feedback, Speech based IR",
author = "Koushik Sinha and Geetha Manjunath and Sharma, {Raveesh R.} and Viswanath Gangavaram and A. Pooja and Murugaian, {Deepak R.}",
year = "2012",
month = "5",
day = "21",
doi = "10.1145/2187980.2188147",
language = "English",
isbn = "9781450312301",
pages = "601--602",
booktitle = "WWW'12 - Proceedings of the 21st Annual Conference on World Wide Web Companion",

}

TY - GEN

T1 - Enabling accent resilient speech based information retrieval

AU - Sinha, Koushik

AU - Manjunath, Geetha

AU - Sharma, Raveesh R.

AU - Gangavaram, Viswanath

AU - Pooja, A.

AU - Murugaian, Deepak R.

PY - 2012/5/21

Y1 - 2012/5/21

N2 - Voice interfaces to browsers and mobile applications are becoming popular as typing with touch screens is cumbersome. The main issue of practical speech based interfaces is how to overcome speech recognition errors. This problem is more severe when the users are non-native speakers of English due to differences in pronunciations. In this paper, we describe a novel, intelligent speech interface design approach for IR tasks that is significantly robust to accent variations. Our solution uses phonemic similarity based word spreading and semantic information based filtering to boost the accuracy of any ASR. We evaluated our solution with Google Voice as the ASR for a web question-answering system developed in-house and the results are very encouraging. Copyright is held by the author/owner(s).

AB - Voice interfaces to browsers and mobile applications are becoming popular as typing with touch screens is cumbersome. The main issue of practical speech based interfaces is how to overcome speech recognition errors. This problem is more severe when the users are non-native speakers of English due to differences in pronunciations. In this paper, we describe a novel, intelligent speech interface design approach for IR tasks that is significantly robust to accent variations. Our solution uses phonemic similarity based word spreading and semantic information based filtering to boost the accuracy of any ASR. We evaluated our solution with Google Voice as the ASR for a web question-answering system developed in-house and the results are very encouraging. Copyright is held by the author/owner(s).

KW - Phonemic spreading

KW - QA

KW - Semantic feedback

KW - Speech based IR

UR - http://www.scopus.com/inward/record.url?scp=84861028464&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84861028464&partnerID=8YFLogxK

U2 - 10.1145/2187980.2188147

DO - 10.1145/2187980.2188147

M3 - Conference contribution

SN - 9781450312301

SP - 601

EP - 602

BT - WWW'12 - Proceedings of the 21st Annual Conference on World Wide Web Companion

ER -