Advances in phonetic word spotting

Arnon Amir, Alon Efrat, Savitha Srinivasan

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

19 Citations (Scopus)

Abstract

Phonetic speech retrieval is used to augment word based retrieval in spoken document retrieval systems, for in and out of vocabulary words. In this paper, we present a new indexing and ranking scheme using metaphones and a Bayesian phonetic edit distance. We conduct an extensive set of experiments using a hundred hours of HUB4 data with ground truth transcript and twenty-four thousands query words. We show improvement of up to 15% in precision compare to results obtained speech recognition alone, at a processing time of 0.5 Sec per query.

Original languageEnglish (US)
Title of host publicationInternational Conference on Information and Knowledge Management, Proceedings
EditorsH. Paques, L. Liu
Pages580-582
Number of pages3
StatePublished - 2001
Externally publishedYes
EventProceedings of the 2001 ACM CIKM: 10th International Conference on Information and Knowledge Management - Atlanta, GA, United States
Duration: Nov 5 2001Nov 10 2001

Other

OtherProceedings of the 2001 ACM CIKM: 10th International Conference on Information and Knowledge Management
CountryUnited States
CityAtlanta, GA
Period11/5/0111/10/01

Fingerprint

Query
Speech recognition
Indexing
Ranking
Experiment

ASJC Scopus subject areas

  • Business, Management and Accounting(all)

Cite this

Amir, A., Efrat, A., & Srinivasan, S. (2001). Advances in phonetic word spotting. In H. Paques, & L. Liu (Eds.), International Conference on Information and Knowledge Management, Proceedings (pp. 580-582)

Advances in phonetic word spotting. / Amir, Arnon; Efrat, Alon; Srinivasan, Savitha.

International Conference on Information and Knowledge Management, Proceedings. ed. / H. Paques; L. Liu. 2001. p. 580-582.

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

Amir, A, Efrat, A & Srinivasan, S 2001, Advances in phonetic word spotting. in H Paques & L Liu (eds), International Conference on Information and Knowledge Management, Proceedings. pp. 580-582, Proceedings of the 2001 ACM CIKM: 10th International Conference on Information and Knowledge Management, Atlanta, GA, United States, 11/5/01.
Amir A, Efrat A, Srinivasan S. Advances in phonetic word spotting. In Paques H, Liu L, editors, International Conference on Information and Knowledge Management, Proceedings. 2001. p. 580-582
Amir, Arnon ; Efrat, Alon ; Srinivasan, Savitha. / Advances in phonetic word spotting. International Conference on Information and Knowledge Management, Proceedings. editor / H. Paques ; L. Liu. 2001. pp. 580-582
@inproceedings{bc5361fd32d04edab55d4de80d0ed91d,
title = "Advances in phonetic word spotting",
abstract = "Phonetic speech retrieval is used to augment word based retrieval in spoken document retrieval systems, for in and out of vocabulary words. In this paper, we present a new indexing and ranking scheme using metaphones and a Bayesian phonetic edit distance. We conduct an extensive set of experiments using a hundred hours of HUB4 data with ground truth transcript and twenty-four thousands query words. We show improvement of up to 15{\%} in precision compare to results obtained speech recognition alone, at a processing time of 0.5 Sec per query.",
author = "Arnon Amir and Alon Efrat and Savitha Srinivasan",
year = "2001",
language = "English (US)",
pages = "580--582",
editor = "H. Paques and L. Liu",
booktitle = "International Conference on Information and Knowledge Management, Proceedings",

}

TY - GEN

T1 - Advances in phonetic word spotting

AU - Amir, Arnon

AU - Efrat, Alon

AU - Srinivasan, Savitha

PY - 2001

Y1 - 2001

N2 - Phonetic speech retrieval is used to augment word based retrieval in spoken document retrieval systems, for in and out of vocabulary words. In this paper, we present a new indexing and ranking scheme using metaphones and a Bayesian phonetic edit distance. We conduct an extensive set of experiments using a hundred hours of HUB4 data with ground truth transcript and twenty-four thousands query words. We show improvement of up to 15% in precision compare to results obtained speech recognition alone, at a processing time of 0.5 Sec per query.

AB - Phonetic speech retrieval is used to augment word based retrieval in spoken document retrieval systems, for in and out of vocabulary words. In this paper, we present a new indexing and ranking scheme using metaphones and a Bayesian phonetic edit distance. We conduct an extensive set of experiments using a hundred hours of HUB4 data with ground truth transcript and twenty-four thousands query words. We show improvement of up to 15% in precision compare to results obtained speech recognition alone, at a processing time of 0.5 Sec per query.

UR - http://www.scopus.com/inward/record.url?scp=0035747557&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035747557&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0035747557

SP - 580

EP - 582

BT - International Conference on Information and Knowledge Management, Proceedings

A2 - Paques, H.

A2 - Liu, L.

ER -