POST: Parallel object-oriented speech toolkit

Jean Hennebert; Dijana Petrovska Delacretaz

POST: Parallel object-oriented speech toolkit

Jean Hennebert
, Dijana Petrovska Delacretaz

ENAC-IIC-GEL

Résultats de recherche: Contribution à une conférence › Papier › Revue par des pairs

2 Citations (Scopus)

Résumé

We give a short overview of POST, a parallel speech toolkit that is distributed freeware to academic institutions. The underlying idea of POST is that large computational problems, like the ones involved in Automatic Speech Recognition (ASR), can be solved more cost effectively by using the aggregate power and memory of many computers. In its current version (January 96) and amongst other things, POST can perform simple feature extraction, training and testing of word and subword Hidden Markov Models (HMMs) with discrete and multigaussian statistical modelling. In this paper, the implementation of the parallelism is discussed and an evaluation of the performances on a telephone database is presented. A short introduction to Parallel Virtual Machine (PVM), the library through which the parallelism is achieved, is also given.

langue originale	Anglais
Pages	1966-1969
Nombre de pages	4
état	Publié - 1 déc. 1996
Modification externe	Oui
Evénement	Proceedings of the 1996 International Conference on Spoken Language Processing, ICSLP. Part 1 (of 4) - Philadelphia, PA, USA Durée: 3 oct. 1996 → 6 oct. 1996

Une conférence

Une conférence	Proceedings of the 1996 International Conference on Spoken Language Processing, ICSLP. Part 1 (of 4)
La ville	Philadelphia, PA, USA
période	3/10/96 → 6/10/96

Contient cette citation

@conference{8af7a4d6f62f4846a88b8f0dd2f79a1e,

title = "POST: Parallel object-oriented speech toolkit",

abstract = "We give a short overview of POST, a parallel speech toolkit that is distributed freeware to academic institutions. The underlying idea of POST is that large computational problems, like the ones involved in Automatic Speech Recognition (ASR), can be solved more cost effectively by using the aggregate power and memory of many computers. In its current version (January 96) and amongst other things, POST can perform simple feature extraction, training and testing of word and subword Hidden Markov Models (HMMs) with discrete and multigaussian statistical modelling. In this paper, the implementation of the parallelism is discussed and an evaluation of the performances on a telephone database is presented. A short introduction to Parallel Virtual Machine (PVM), the library through which the parallelism is achieved, is also given.",

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

year = "1996",

month = dec,

day = "1",

language = "English",

pages = "1966--1969",

note = "Proceedings of the 1996 International Conference on Spoken Language Processing, ICSLP. Part 1 (of 4) ; Conference date: 03-10-1996 Through 06-10-1996",

}

TY - CONF

T1 - POST

T2 - Proceedings of the 1996 International Conference on Spoken Language Processing, ICSLP. Part 1 (of 4)

AU - Hennebert, Jean

AU - Delacretaz, Dijana Petrovska

PY - 1996/12/1

Y1 - 1996/12/1

N2 - We give a short overview of POST, a parallel speech toolkit that is distributed freeware to academic institutions. The underlying idea of POST is that large computational problems, like the ones involved in Automatic Speech Recognition (ASR), can be solved more cost effectively by using the aggregate power and memory of many computers. In its current version (January 96) and amongst other things, POST can perform simple feature extraction, training and testing of word and subword Hidden Markov Models (HMMs) with discrete and multigaussian statistical modelling. In this paper, the implementation of the parallelism is discussed and an evaluation of the performances on a telephone database is presented. A short introduction to Parallel Virtual Machine (PVM), the library through which the parallelism is achieved, is also given.

AB - We give a short overview of POST, a parallel speech toolkit that is distributed freeware to academic institutions. The underlying idea of POST is that large computational problems, like the ones involved in Automatic Speech Recognition (ASR), can be solved more cost effectively by using the aggregate power and memory of many computers. In its current version (January 96) and amongst other things, POST can perform simple feature extraction, training and testing of word and subword Hidden Markov Models (HMMs) with discrete and multigaussian statistical modelling. In this paper, the implementation of the parallelism is discussed and an evaluation of the performances on a telephone database is presented. A short introduction to Parallel Virtual Machine (PVM), the library through which the parallelism is achieved, is also given.

M3 - Paper

AN - SCOPUS:0030376676

SP - 1966

EP - 1969

Y2 - 3 October 1996 through 6 October 1996

ER -

POST: Parallel object-oriented speech toolkit

Résumé

Une conférence

Empreinte digitale

Contient cette citation