Text-prompted speaker verification experiments with phoneme specific MLPs

Dijana Petrovska Delacretaz; Jean Hennebert

doi:10.1109/ICASSP.1998.675380

Text-prompted speaker verification experiments with phoneme specific MLPs

Dijana Petrovska Delacretaz
, Jean Hennebert

ETH
Circuits and Systems Group

Résultats de recherche: Le chapitre dans un livre, un rapport, une anthologie ou une collection › Contribution à une conférence › Revue par des pairs

28 Citations (Scopus)

Résumé

The aims of the study described in this paper are (1) to assess the relative speaker discriminant properties of phonemes and (2) to investigate the importance of the temporal frame-to-frame information for speaker modelling in the framework of a text-prompted speaker verification system using hidden Markov models (HMMs) and multilayer perceptrons (MLPs). It is shown that, with similar experimental conditions, nasals, fricatives and vowels convey more speaker specific information than plosives and liquids. Regarding the influence of the frame-to-frame temporal information, significant improvements are reported from the inclusion of several acoustic frames at the input of the MLPs. The results tend also to show that each phoneme has its optimal MLP context size giving the best equal error rate (EER).

langue originale	Anglais
titre	Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998
Pages	777-780
Nombre de pages	4
Les DOIs	https://doi.org/10.1109/ICASSP.1998.675380
état	Publié - 1 déc. 1998
Modification externe	Oui
Evénement	1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998 - Seattle, WA, États-Unis Durée: 12 mai 1998 → 15 mai 1998

Série de publications

Nom	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2
ISSN (imprimé)	1520-6149

Une conférence

Une conférence	1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998
Pays/Territoire	États-Unis
La ville	Seattle, WA
période	12/05/98 → 15/05/98

Accès au document

10.1109/ICASSP.1998.675380

Contient cette citation

Delacretaz, D. P., & Hennebert, J. (1998). Text-prompted speaker verification experiments with phoneme specific MLPs. Dans Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998 (p. 777-780). Article 675380 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol 2). https://doi.org/10.1109/ICASSP.1998.675380

@inproceedings{fdca958e39094bb8a47176f78a57bc52,

title = "Text-prompted speaker verification experiments with phoneme specific MLPs",

abstract = "The aims of the study described in this paper are (1) to assess the relative speaker discriminant properties of phonemes and (2) to investigate the importance of the temporal frame-to-frame information for speaker modelling in the framework of a text-prompted speaker verification system using hidden Markov models (HMMs) and multilayer perceptrons (MLPs). It is shown that, with similar experimental conditions, nasals, fricatives and vowels convey more speaker specific information than plosives and liquids. Regarding the influence of the frame-to-frame temporal information, significant improvements are reported from the inclusion of several acoustic frames at the input of the MLPs. The results tend also to show that each phoneme has its optimal MLP context size giving the best equal error rate (EER).",

author = "Delacretaz, \{Dijana Petrovska\} and Jean Hennebert",

year = "1998",

month = dec,

day = "1",

doi = "10.1109/ICASSP.1998.675380",

language = "English",

isbn = "0780344286",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

pages = "777--780",

booktitle = "Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998",

note = "1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998 ; Conference date: 12-05-1998 Through 15-05-1998",

}

Delacretaz, DP & Hennebert, J 1998, Text-prompted speaker verification experiments with phoneme specific MLPs. Dans Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998., 675380, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, VOL. 2, p. 777-780, 1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998, Seattle, WA, États-Unis, 12/05/98. https://doi.org/10.1109/ICASSP.1998.675380

Text-prompted speaker verification experiments with phoneme specific MLPs. / Delacretaz, Dijana Petrovska; Hennebert, Jean.
Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998. 1998. p. 777-780 675380 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol 2).

Résultats de recherche: Le chapitre dans un livre, un rapport, une anthologie ou une collection › Contribution à une conférence › Revue par des pairs

TY - GEN

T1 - Text-prompted speaker verification experiments with phoneme specific MLPs

AU - Delacretaz, Dijana Petrovska

AU - Hennebert, Jean

PY - 1998/12/1

Y1 - 1998/12/1

N2 - The aims of the study described in this paper are (1) to assess the relative speaker discriminant properties of phonemes and (2) to investigate the importance of the temporal frame-to-frame information for speaker modelling in the framework of a text-prompted speaker verification system using hidden Markov models (HMMs) and multilayer perceptrons (MLPs). It is shown that, with similar experimental conditions, nasals, fricatives and vowels convey more speaker specific information than plosives and liquids. Regarding the influence of the frame-to-frame temporal information, significant improvements are reported from the inclusion of several acoustic frames at the input of the MLPs. The results tend also to show that each phoneme has its optimal MLP context size giving the best equal error rate (EER).

AB - The aims of the study described in this paper are (1) to assess the relative speaker discriminant properties of phonemes and (2) to investigate the importance of the temporal frame-to-frame information for speaker modelling in the framework of a text-prompted speaker verification system using hidden Markov models (HMMs) and multilayer perceptrons (MLPs). It is shown that, with similar experimental conditions, nasals, fricatives and vowels convey more speaker specific information than plosives and liquids. Regarding the influence of the frame-to-frame temporal information, significant improvements are reported from the inclusion of several acoustic frames at the input of the MLPs. The results tend also to show that each phoneme has its optimal MLP context size giving the best equal error rate (EER).

U2 - 10.1109/ICASSP.1998.675380

DO - 10.1109/ICASSP.1998.675380

M3 - Conference contribution

AN - SCOPUS:9444269409

SN - 0780344286

SN - 9780780344280

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 777

EP - 780

BT - Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998

T2 - 1998 23rd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998

Y2 - 12 May 1998 through 15 May 1998

ER -

Delacretaz DP, Hennebert J. Text-prompted speaker verification experiments with phoneme specific MLPs. Dans Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998. 1998. p. 777-780. 675380. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.1998.675380

Text-prompted speaker verification experiments with phoneme specific MLPs

Résumé

Série de publications

Une conférence

Accès au document

Empreinte digitale

Contient cette citation