Speaker Identification Using Glottal-Source Waveforms and Support-Vector-Machine Modelling

David Vandyke, Michael Wagner, Girija Chetty, Roland Goecke

Research output: A Conference proceeding or a Chapter in BookConference contribution

Abstract

Speaker identification experiments are performed with novel features representative of the glottal source waveform. These are derived from closed-phase analysis and inverse filtering. Source waveforms are segmented into two consecutive periods and normalised in prosody, forming so called source-frame feature vectors. Support-vector-machines are used to construct speaker discriminative hyperplanes and identification rates are reported. Groups of male speakers of size 5 to 20 are examined from the YOHO corpus and 65% correct identification rates are achieved on a per source-frame basis. Finally the source-frames phonetic independence is confirmed with the TI 46-Word corpus.
Original languageEnglish
Title of host publicationProceedings of the 14th Australasian International Conference on Speech Science and Technology
EditorsFelicity Cox, Katherine Demuth, Susan Lin, Kelly Miles, Sallyanne Palethrope, Jason Shaw, Ivan Yuen
Place of PublicationSydney, Australia
PublisherAustralasian Speech Science and Technology Association (ASSTA)
Pages49-52
Number of pages4
Publication statusPublished - 2012
Event14th Australasian International Conference on Speech Science and Technology - Sydney, Sydney, Australia
Duration: 3 Dec 20126 Dec 2012

Publication series

NameASSTA 2013
PublisherAustralasian Speech Science and Technology Association
ISSN (Print)1039-0227

Conference

Conference14th Australasian International Conference on Speech Science and Technology
CountryAustralia
CitySydney
Period3/12/126/12/12

Fingerprint

Speech analysis
Support vector machines
Experiments

Cite this

Vandyke, D., Wagner, M., Chetty, G., & Goecke, R. (2012). Speaker Identification Using Glottal-Source Waveforms and Support-Vector-Machine Modelling. In F. Cox, K. Demuth, S. Lin, K. Miles, S. Palethrope, J. Shaw, & I. Yuen (Eds.), Proceedings of the 14th Australasian International Conference on Speech Science and Technology (pp. 49-52). (ASSTA 2013). Sydney, Australia: Australasian Speech Science and Technology Association (ASSTA).
Vandyke, David ; Wagner, Michael ; Chetty, Girija ; Goecke, Roland. / Speaker Identification Using Glottal-Source Waveforms and Support-Vector-Machine Modelling. Proceedings of the 14th Australasian International Conference on Speech Science and Technology. editor / Felicity Cox ; Katherine Demuth ; Susan Lin ; Kelly Miles ; Sallyanne Palethrope ; Jason Shaw ; Ivan Yuen. Sydney, Australia : Australasian Speech Science and Technology Association (ASSTA), 2012. pp. 49-52 (ASSTA 2013).
@inproceedings{dafcbe462e004c95954e0c8b7d6b46d7,
title = "Speaker Identification Using Glottal-Source Waveforms and Support-Vector-Machine Modelling",
abstract = "Speaker identification experiments are performed with novel features representative of the glottal source waveform. These are derived from closed-phase analysis and inverse filtering. Source waveforms are segmented into two consecutive periods and normalised in prosody, forming so called source-frame feature vectors. Support-vector-machines are used to construct speaker discriminative hyperplanes and identification rates are reported. Groups of male speakers of size 5 to 20 are examined from the YOHO corpus and 65{\%} correct identification rates are achieved on a per source-frame basis. Finally the source-frames phonetic independence is confirmed with the TI 46-Word corpus.",
keywords = "Speaker identification, Glottal Source waveform, Support vector machine, Linear Predictive Coding",
author = "David Vandyke and Michael Wagner and Girija Chetty and Roland Goecke",
year = "2012",
language = "English",
series = "ASSTA 2013",
publisher = "Australasian Speech Science and Technology Association (ASSTA)",
pages = "49--52",
editor = "Felicity Cox and Katherine Demuth and Susan Lin and Kelly Miles and Sallyanne Palethrope and Jason Shaw and Ivan Yuen",
booktitle = "Proceedings of the 14th Australasian International Conference on Speech Science and Technology",

}

Vandyke, D, Wagner, M, Chetty, G & Goecke, R 2012, Speaker Identification Using Glottal-Source Waveforms and Support-Vector-Machine Modelling. in F Cox, K Demuth, S Lin, K Miles, S Palethrope, J Shaw & I Yuen (eds), Proceedings of the 14th Australasian International Conference on Speech Science and Technology. ASSTA 2013, Australasian Speech Science and Technology Association (ASSTA), Sydney, Australia, pp. 49-52, 14th Australasian International Conference on Speech Science and Technology, Sydney, Australia, 3/12/12.

Speaker Identification Using Glottal-Source Waveforms and Support-Vector-Machine Modelling. / Vandyke, David; Wagner, Michael; Chetty, Girija; Goecke, Roland.

Proceedings of the 14th Australasian International Conference on Speech Science and Technology. ed. / Felicity Cox; Katherine Demuth; Susan Lin; Kelly Miles; Sallyanne Palethrope; Jason Shaw; Ivan Yuen. Sydney, Australia : Australasian Speech Science and Technology Association (ASSTA), 2012. p. 49-52 (ASSTA 2013).

Research output: A Conference proceeding or a Chapter in BookConference contribution

TY - GEN

T1 - Speaker Identification Using Glottal-Source Waveforms and Support-Vector-Machine Modelling

AU - Vandyke, David

AU - Wagner, Michael

AU - Chetty, Girija

AU - Goecke, Roland

PY - 2012

Y1 - 2012

N2 - Speaker identification experiments are performed with novel features representative of the glottal source waveform. These are derived from closed-phase analysis and inverse filtering. Source waveforms are segmented into two consecutive periods and normalised in prosody, forming so called source-frame feature vectors. Support-vector-machines are used to construct speaker discriminative hyperplanes and identification rates are reported. Groups of male speakers of size 5 to 20 are examined from the YOHO corpus and 65% correct identification rates are achieved on a per source-frame basis. Finally the source-frames phonetic independence is confirmed with the TI 46-Word corpus.

AB - Speaker identification experiments are performed with novel features representative of the glottal source waveform. These are derived from closed-phase analysis and inverse filtering. Source waveforms are segmented into two consecutive periods and normalised in prosody, forming so called source-frame feature vectors. Support-vector-machines are used to construct speaker discriminative hyperplanes and identification rates are reported. Groups of male speakers of size 5 to 20 are examined from the YOHO corpus and 65% correct identification rates are achieved on a per source-frame basis. Finally the source-frames phonetic independence is confirmed with the TI 46-Word corpus.

KW - Speaker identification

KW - Glottal Source waveform

KW - Support vector machine

KW - Linear Predictive Coding

UR - http://assta.org/sst/SST-12/SST2012/PDF/INDEXSCR.PDF#toc

M3 - Conference contribution

T3 - ASSTA 2013

SP - 49

EP - 52

BT - Proceedings of the 14th Australasian International Conference on Speech Science and Technology

A2 - Cox, Felicity

A2 - Demuth, Katherine

A2 - Lin, Susan

A2 - Miles, Kelly

A2 - Palethrope, Sallyanne

A2 - Shaw, Jason

A2 - Yuen, Ivan

PB - Australasian Speech Science and Technology Association (ASSTA)

CY - Sydney, Australia

ER -

Vandyke D, Wagner M, Chetty G, Goecke R. Speaker Identification Using Glottal-Source Waveforms and Support-Vector-Machine Modelling. In Cox F, Demuth K, Lin S, Miles K, Palethrope S, Shaw J, Yuen I, editors, Proceedings of the 14th Australasian International Conference on Speech Science and Technology. Sydney, Australia: Australasian Speech Science and Technology Association (ASSTA). 2012. p. 49-52. (ASSTA 2013).