Large population speaker identification using clean and telephone speech

Author

Reynolds, Douglas A.

Author_Institution

Lincoln Lab., MIT, Cambridge, MA, USA

Volume

2

Issue

3

fYear

1995

fDate

3/1/1995 12:00:00 AM

Firstpage

46

Lastpage

48

Abstract

This paper presents text-independent speaker identification results for varying speaker population sizes up to 630 speakers for both clean, wideband speech, and telephone speech. A system based on Gaussian mixture speaker models is used for speaker identification, and experiments are conducted on the TIMIT and NTIMIT databases. The TIMIT results show large population performance under near-ideal conditions, and the NTIMIT results show the corresponding accuracy loss due to telephone transmission. These are believed to be the first speaker identification experiments on the complete 630 speaker TIMIT and NTIMIT databases and the largest text-independent speaker identification task reported to date. Identification accuracies of 99.5 and 60.7% were achieved on the TIMIT and NTIMIT databases, respectively.<>

Keywords

Gaussian processes; speaker recognition; telephony; Gaussian mixture speaker models; NTIMIT database; TIMIT database; clean speech; large population speaker identification; near-ideal conditions; telephone speech; telephone transmission; text-independent speaker identification; wideband speech; Additive noise; Databases; Degradation; Loudspeakers; Performance loss; Propagation losses; Speech analysis; Telephony; Wideband;

fLanguage

English

Journal_Title

Signal Processing Letters, IEEE

Publisher

ieee

ISSN

1070-9908

Type

jour

DOI

10.1109/97.372913

Filename

372913