Combining transcription-based and acoustic-based speaker identifications for broadcast news

Author

El Khoury, Elie ; Laurent, Antoine ; Meignier, Sylvain ; Petitrenaud, Simon

Author_Institution

LIUM, Univ. du Maine-Le Mans, Le Mans, France

fYear

2012

fDate

25-30 March 2012

Firstpage

4377

Lastpage

4380

Abstract

In this paper, we consider the issue of speaker identification within audio records of broadcast news. The speaker identity information is extracted from both transcript-based and acoustic-based speaker identification systems. This information is combined in the belief functions framework, which makes coherent the knowledge representation of the problem. The Kuhn-Munkres algorithm is used to optimize the assignment problem of speaker identities and speaker clusters. Experiments carried out on French broadcast news from the French evaluation campaign ESTER show the efficiency of the proposed combination method.

Keywords

audio recording; belief networks; pattern clustering; speaker recognition; French broadcast news; French evaluation campaign ESTER; Kuhn-Munkres algorithm; acoustic-based speaker identifications; audio records; belief functions framework; knowledge representation; speaker clusters; speaker identities; speaker identity information; transcription-based speaker identifications; Acoustics; Computational modeling; Decoding; Error analysis; Hidden Markov models; Mathematical model; Training; Speaker identification; belief functions; speaker diarization;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing (ICASSP), 2012 IEEE International Conference on

Conference_Location

Kyoto

ISSN

1520-6149

Print_ISBN

978-1-4673-0045-2

Electronic_ISBN

1520-6149

Type

conf

DOI

10.1109/ICASSP.2012.6288889

Filename

6288889