Phonetic name matching for cross-lingual spoken sentence retrieval

Heng Ji; Ralph Grishman; Wen Wang

doi:10.1109/SLT.2008.4777895

Phonetic name matching for cross-lingual spoken sentence retrieval

Heng Ji, Ralph Grishman, Wen Wang

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Cross-lingual Spoken Sentence Retrieval (CLSSR) remains a challenge, especially for queries including OOV words such as person names. This paper proposes a simple method of fuzzy matching between query names and phones of candidate audio segments. This approach has the advantage of avoiding some word decoding errors in Automatic Speech Recognition (ASR). Experiments on Mandarin-English CLSSR show that phone-based searching and conventional translation-based searching are complementary. Adding phone matching achieved 26.29% improvement on F-measure over searching on state-of-the-art Machine Translation (MT) output and 8.83% over Entity Translation (ET) output.

Original language	English (US)
Title of host publication	2008 IEEE Workshop on Spoken Language Technology, SLT 2008 - Proceedings
Pages	281-284
Number of pages	4
DOIs	https://doi.org/10.1109/SLT.2008.4777895
State	Published - 2008
Event	2008 IEEE Workshop on Spoken Language Technology, SLT 2008 - Goa, India Duration: Dec 15 2008 → Dec 19 2008

Publication series

Name	2008 IEEE Workshop on Spoken Language Technology, SLT 2008 - Proceedings

Other

Other	2008 IEEE Workshop on Spoken Language Technology, SLT 2008
Country/Territory	India
City	Goa
Period	12/15/08 → 12/19/08

Keywords

Speech Recognition, Information Retrieval

ASJC Scopus subject areas

Language and Linguistics
Software
Electrical and Electronic Engineering
Communication

Access to Document

10.1109/SLT.2008.4777895

Cite this

Phonetic name matching for cross-lingual spoken sentence retrieval. / Ji, Heng; Grishman, Ralph; Wang, Wen.
2008 IEEE Workshop on Spoken Language Technology, SLT 2008 - Proceedings. 2008. p. 281-284 4777895 (2008 IEEE Workshop on Spoken Language Technology, SLT 2008 - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Ji, H, Grishman, R & Wang, W 2008, Phonetic name matching for cross-lingual spoken sentence retrieval. in 2008 IEEE Workshop on Spoken Language Technology, SLT 2008 - Proceedings., 4777895, 2008 IEEE Workshop on Spoken Language Technology, SLT 2008 - Proceedings, pp. 281-284, 2008 IEEE Workshop on Spoken Language Technology, SLT 2008, Goa, India, 12/15/08. https://doi.org/10.1109/SLT.2008.4777895

@inproceedings{b5598c236d6a42798b4b2f4bd34d35d0,

title = "Phonetic name matching for cross-lingual spoken sentence retrieval",

abstract = "Cross-lingual Spoken Sentence Retrieval (CLSSR) remains a challenge, especially for queries including OOV words such as person names. This paper proposes a simple method of fuzzy matching between query names and phones of candidate audio segments. This approach has the advantage of avoiding some word decoding errors in Automatic Speech Recognition (ASR). Experiments on Mandarin-English CLSSR show that phone-based searching and conventional translation-based searching are complementary. Adding phone matching achieved 26.29% improvement on F-measure over searching on state-of-the-art Machine Translation (MT) output and 8.83% over Entity Translation (ET) output.",

keywords = "Speech Recognition, Information Retrieval",

author = "Heng Ji and Ralph Grishman and Wen Wang",

year = "2008",

doi = "10.1109/SLT.2008.4777895",

language = "English (US)",

isbn = "9781424434725",

series = "2008 IEEE Workshop on Spoken Language Technology, SLT 2008 - Proceedings",

pages = "281--284",

booktitle = "2008 IEEE Workshop on Spoken Language Technology, SLT 2008 - Proceedings",

note = "2008 IEEE Workshop on Spoken Language Technology, SLT 2008 ; Conference date: 15-12-2008 Through 19-12-2008",

}

TY - GEN

T1 - Phonetic name matching for cross-lingual spoken sentence retrieval

AU - Ji, Heng

AU - Grishman, Ralph

AU - Wang, Wen

PY - 2008

Y1 - 2008

N2 - Cross-lingual Spoken Sentence Retrieval (CLSSR) remains a challenge, especially for queries including OOV words such as person names. This paper proposes a simple method of fuzzy matching between query names and phones of candidate audio segments. This approach has the advantage of avoiding some word decoding errors in Automatic Speech Recognition (ASR). Experiments on Mandarin-English CLSSR show that phone-based searching and conventional translation-based searching are complementary. Adding phone matching achieved 26.29% improvement on F-measure over searching on state-of-the-art Machine Translation (MT) output and 8.83% over Entity Translation (ET) output.

AB - Cross-lingual Spoken Sentence Retrieval (CLSSR) remains a challenge, especially for queries including OOV words such as person names. This paper proposes a simple method of fuzzy matching between query names and phones of candidate audio segments. This approach has the advantage of avoiding some word decoding errors in Automatic Speech Recognition (ASR). Experiments on Mandarin-English CLSSR show that phone-based searching and conventional translation-based searching are complementary. Adding phone matching achieved 26.29% improvement on F-measure over searching on state-of-the-art Machine Translation (MT) output and 8.83% over Entity Translation (ET) output.

KW - Speech Recognition, Information Retrieval

UR - http://www.scopus.com/inward/record.url?scp=67649552768&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67649552768&partnerID=8YFLogxK

U2 - 10.1109/SLT.2008.4777895

DO - 10.1109/SLT.2008.4777895

M3 - Conference contribution

AN - SCOPUS:67649552768

SN - 9781424434725

T3 - 2008 IEEE Workshop on Spoken Language Technology, SLT 2008 - Proceedings

SP - 281

EP - 284

BT - 2008 IEEE Workshop on Spoken Language Technology, SLT 2008 - Proceedings

T2 - 2008 IEEE Workshop on Spoken Language Technology, SLT 2008

Y2 - 15 December 2008 through 19 December 2008

ER -

Phonetic name matching for cross-lingual spoken sentence retrieval

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this