Australian accent-based speaker classification

Dat Tran; Xu Huang; Dharmendra Sharma

doi:10.1109/WKDD.2010.80

Australian accent-based speaker classification

Dat Tran, Xu Huang, Dharmendra Sharma

Information, Technology (IT) & Systems

Research output: A Conference proceeding or a Chapter in Book › Conference contribution › peer-review

8 Citations (Scopus)

1034 Downloads (Pure)

Abstract

An automatic voice-based classification system of speaker characteristics including age, gender and accent is presented in this paper. Speakers are grouped according to their characteristics and their speech features are then extracted to train speaker group models using different classification techniques. Finally fusion of classification results for those speaker groups is performed to obtain results for each speaker characteristic. The ANDOSL Australian speech database consisting of 108 speakers and 21600 long utterances was used for system evaluation. Experiments showed high performance for the proposed classification of speaker characteristics.

Original language	English
Title of host publication	2010 Third International Conference on Knowledge Discovery and Data Mining: WKDD 2010
Place of Publication	Piscataway, N.J., USA
Publisher	IEEE
Pages	416-419
Number of pages	4
ISBN (Print)	9781424470556
DOIs	https://doi.org/10.1109/WKDD.2010.80
Publication status	Published - 2010
Event	2010 The 3rd International Conference on Knowledge Discovery and Data Mining (WKDD 2010) - Phuket, Thailand Duration: 9 Jan 2010 → 10 Jan 2010

Conference

Conference	2010 The 3rd International Conference on Knowledge Discovery and Data Mining (WKDD 2010)
Country/Territory	Thailand
City	Phuket
Period	9/01/10 → 10/01/10

Access to Document

10.1109/WKDD.2010.80

PaperFinal published version, 264 KB

Cite this

@inproceedings{af7fbf1380d24aee96713858c5bea23e,

title = "Australian accent-based speaker classification",

abstract = "An automatic voice-based classification system of speaker characteristics including age, gender and accent is presented in this paper. Speakers are grouped according to their characteristics and their speech features are then extracted to train speaker group models using different classification techniques. Finally fusion of classification results for those speaker groups is performed to obtain results for each speaker characteristic. The ANDOSL Australian speech database consisting of 108 speakers and 21600 long utterances was used for system evaluation. Experiments showed high performance for the proposed classification of speaker characteristics.",

author = "Dat Tran and Xu Huang and Dharmendra Sharma",

year = "2010",

doi = "10.1109/WKDD.2010.80",

language = "English",

isbn = "9781424470556",

pages = "416--419",

booktitle = "2010 Third International Conference on Knowledge Discovery and Data Mining: WKDD 2010",

publisher = "IEEE",

note = "2010 The 3rd International Conference on Knowledge Discovery and Data Mining (WKDD 2010) ; Conference date: 09-01-2010 Through 10-01-2010",

}

TY - GEN

T1 - Australian accent-based speaker classification

AU - Tran, Dat

AU - Huang, Xu

AU - Sharma, Dharmendra

PY - 2010

Y1 - 2010

N2 - An automatic voice-based classification system of speaker characteristics including age, gender and accent is presented in this paper. Speakers are grouped according to their characteristics and their speech features are then extracted to train speaker group models using different classification techniques. Finally fusion of classification results for those speaker groups is performed to obtain results for each speaker characteristic. The ANDOSL Australian speech database consisting of 108 speakers and 21600 long utterances was used for system evaluation. Experiments showed high performance for the proposed classification of speaker characteristics.

AB - An automatic voice-based classification system of speaker characteristics including age, gender and accent is presented in this paper. Speakers are grouped according to their characteristics and their speech features are then extracted to train speaker group models using different classification techniques. Finally fusion of classification results for those speaker groups is performed to obtain results for each speaker characteristic. The ANDOSL Australian speech database consisting of 108 speakers and 21600 long utterances was used for system evaluation. Experiments showed high performance for the proposed classification of speaker characteristics.

U2 - 10.1109/WKDD.2010.80

DO - 10.1109/WKDD.2010.80

M3 - Conference contribution

SN - 9781424470556

SP - 416

EP - 419

BT - 2010 Third International Conference on Knowledge Discovery and Data Mining: WKDD 2010

PB - IEEE

CY - Piscataway, N.J., USA

T2 - 2010 The 3rd International Conference on Knowledge Discovery and Data Mining (WKDD 2010)

Y2 - 9 January 2010 through 10 January 2010

ER -

Australian accent-based speaker classification

Abstract

Conference

Access to Document

Fingerprint

Cite this