Unsupervised speaker classification using self-organizing maps (SOM)

Itshak Voitovetsky; Hugo Guterman; Arnon Cohen

Unsupervised speaker classification using self-organizing maps (SOM)

Itshak Voitovetsky, Hugo Guterman, Arnon Cohen

Department of Electrical & Computer Engineering

Research output: Contribution to conference › Paper › peer-review

12 Scopus citations

Abstract

An algorithm for unsupervised speaker classification using Kohonen SOM is presented. The system employs 6×10 SOM networks for each speaker and for non-speech segments. The algorithm was evaluated using high quality as well as telephone quality conversations between two speakers. Correct classification of more than 90% was demonstrated. High quality conversation between three speakers yielded 80% correct classification. The high quality speech required the use of 12^th order cepstral coefficients vector. In telephone quality speech, additional 12 features of the difference of the cepstrum were required.

Original language	English
Pages	578-587
Number of pages	10
State	Published - 1 Dec 1997
Event	Proceedings of the 1997 7th IEEE Workshop on Neural Networks for Signal Processing, NNSP'97 - Amelia Island, FL, USA Duration: 24 Sep 1997 → 26 Sep 1997

Conference

Conference	Proceedings of the 1997 7th IEEE Workshop on Neural Networks for Signal Processing, NNSP'97
City	Amelia Island, FL, USA
Period	24/09/97 → 26/09/97

ASJC Scopus subject areas

Signal Processing
Software
Electrical and Electronic Engineering

Cite this

@conference{8ea8528ee0114e05ae5bd9f3c577ef83,

title = "Unsupervised speaker classification using self-organizing maps (SOM)",

abstract = "An algorithm for unsupervised speaker classification using Kohonen SOM is presented. The system employs 6×10 SOM networks for each speaker and for non-speech segments. The algorithm was evaluated using high quality as well as telephone quality conversations between two speakers. Correct classification of more than 90% was demonstrated. High quality conversation between three speakers yielded 80% correct classification. The high quality speech required the use of 12th order cepstral coefficients vector. In telephone quality speech, additional 12 features of the difference of the cepstrum were required.",

author = "Itshak Voitovetsky and Hugo Guterman and Arnon Cohen",

year = "1997",

month = dec,

day = "1",

language = "English",

pages = "578--587",

note = "Proceedings of the 1997 7th IEEE Workshop on Neural Networks for Signal Processing, NNSP'97 ; Conference date: 24-09-1997 Through 26-09-1997",

}

TY - CONF

T1 - Unsupervised speaker classification using self-organizing maps (SOM)

AU - Voitovetsky, Itshak

AU - Guterman, Hugo

AU - Cohen, Arnon

PY - 1997/12/1

Y1 - 1997/12/1

N2 - An algorithm for unsupervised speaker classification using Kohonen SOM is presented. The system employs 6×10 SOM networks for each speaker and for non-speech segments. The algorithm was evaluated using high quality as well as telephone quality conversations between two speakers. Correct classification of more than 90% was demonstrated. High quality conversation between three speakers yielded 80% correct classification. The high quality speech required the use of 12th order cepstral coefficients vector. In telephone quality speech, additional 12 features of the difference of the cepstrum were required.

AB - An algorithm for unsupervised speaker classification using Kohonen SOM is presented. The system employs 6×10 SOM networks for each speaker and for non-speech segments. The algorithm was evaluated using high quality as well as telephone quality conversations between two speakers. Correct classification of more than 90% was demonstrated. High quality conversation between three speakers yielded 80% correct classification. The high quality speech required the use of 12th order cepstral coefficients vector. In telephone quality speech, additional 12 features of the difference of the cepstrum were required.

UR - http://www.scopus.com/inward/record.url?scp=0031331636&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0031331636

SP - 578

EP - 587

T2 - Proceedings of the 1997 7th IEEE Workshop on Neural Networks for Signal Processing, NNSP'97

Y2 - 24 September 1997 through 26 September 1997

ER -

Unsupervised speaker classification using self-organizing maps (SOM)

Abstract

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this