Telephone conversation speaker diarization using mealy-HMMs

Itshak Lapidot; Jean Francois Bonastre; Samy Bengio

Telephone conversation speaker diarization using mealy-HMMs

Itshak Lapidot
, Jean Francois Bonastre
, Samy Bengio

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

1 Scopus citations

Abstract

When Hidden Markov Models (HMMs) were first introduced, two competing representation models were proposed, the Moore model, with separate emission and transition distributions, which is commonly used in speech technologies, and the Mealy model, with a single emission-transition distribution. Since then the literature has mostly focused on the Moore model. In this paper, we would like to show the use of Mealy- HMMs for telephone conversation speaker diarization task. We present the Viterbi training and decoding for Mealy-HMMs and show that it yields similar performance compared to Moore- HMMs with a fewer number of parameters.

Original language	English
Pages	173-178
Number of pages	6
State	Published - 1 Jan 2014
Externally published	Yes
Event	Speaker and Language Recognition Workshop, Odyssey 2014 - Joensuu, Finland Duration: 16 Jun 2014 → 19 Jun 2014

Conference

Conference	Speaker and Language Recognition Workshop, Odyssey 2014
Country/Territory	Finland
City	Joensuu
Period	16/06/14 → 19/06/14

ASJC Scopus subject areas

Signal Processing
Software
Human-Computer Interaction

Cite this

@conference{3f10096e60354fce9343af9c7ecd6034,

title = "Telephone conversation speaker diarization using mealy-HMMs",

abstract = "When Hidden Markov Models (HMMs) were first introduced, two competing representation models were proposed, the Moore model, with separate emission and transition distributions, which is commonly used in speech technologies, and the Mealy model, with a single emission-transition distribution. Since then the literature has mostly focused on the Moore model. In this paper, we would like to show the use of Mealy- HMMs for telephone conversation speaker diarization task. We present the Viterbi training and decoding for Mealy-HMMs and show that it yields similar performance compared to Moore- HMMs with a fewer number of parameters.",

author = "Itshak Lapidot and Bonastre, \{Jean Francois\} and Samy Bengio",

note = "Publisher Copyright: {\textcopyright} Odyssey 2014: Speaker and Language Recognition Workshop. All rights Reserved.; Speaker and Language Recognition Workshop, Odyssey 2014 ; Conference date: 16-06-2014 Through 19-06-2014",

year = "2014",

month = jan,

day = "1",

language = "English",

pages = "173--178",

}

TY - CONF

T1 - Telephone conversation speaker diarization using mealy-HMMs

AU - Lapidot, Itshak

AU - Bonastre, Jean Francois

AU - Bengio, Samy

PY - 2014/1/1

Y1 - 2014/1/1

N2 - When Hidden Markov Models (HMMs) were first introduced, two competing representation models were proposed, the Moore model, with separate emission and transition distributions, which is commonly used in speech technologies, and the Mealy model, with a single emission-transition distribution. Since then the literature has mostly focused on the Moore model. In this paper, we would like to show the use of Mealy- HMMs for telephone conversation speaker diarization task. We present the Viterbi training and decoding for Mealy-HMMs and show that it yields similar performance compared to Moore- HMMs with a fewer number of parameters.

AB - When Hidden Markov Models (HMMs) were first introduced, two competing representation models were proposed, the Moore model, with separate emission and transition distributions, which is commonly used in speech technologies, and the Mealy model, with a single emission-transition distribution. Since then the literature has mostly focused on the Moore model. In this paper, we would like to show the use of Mealy- HMMs for telephone conversation speaker diarization task. We present the Viterbi training and decoding for Mealy-HMMs and show that it yields similar performance compared to Moore- HMMs with a fewer number of parameters.

UR - https://www.scopus.com/pages/publications/85014183981

M3 - Paper

AN - SCOPUS:85014183981

SP - 173

EP - 178

T2 - Speaker and Language Recognition Workshop, Odyssey 2014

Y2 - 16 June 2014 through 19 June 2014

ER -

Telephone conversation speaker diarization using mealy-HMMs

Abstract

Conference

ASJC Scopus subject areas

Fingerprint

Cite this