Hidden neural networks: application to speech recognition

Søren Kamaric Riis

doi:10.1109/ICASSP.1998.675465

Hidden neural networks: application to speech recognition

Søren Kamaric Riis

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

460 Downloads (Orbit)

Abstract

We evaluate the hidden neural network HMM/NN hybrid on two speech recognition benchmark tasks; (1) task independent isolated word recognition on the Phonebook database, and (2) recognition of broad phoneme classes in continuous speech from the TIMIT database. It is shown how hidden neural networks (HNNs) with much fewer parameters than conventional HMMs and other hybrids can obtain comparable performance, and for the broad class task it is illustrated how the HNN can be applied as a purely transition based system, where acoustic context dependent transition probabilities are estimated by neural networks

Original language	English
Title of host publication	Acoustics, Speech and Signal Processing, 1998. Proceedings of the 1998 IEEE International Conference on
Volume	2
Publisher	IEEE
Publication date	1998
Pages	1117-1120
ISBN (Print)	0-7803-4428-6
DOIs	https://doi.org/10.1109/ICASSP.1998.675465
Publication status	Published - 1998
Event	1998 IEEE International Conference on Acoustics, Speech and Signal Processing - Seattle, United States Duration: 12 May 1998 → 15 May 1998 Conference number: 23

Conference

Conference	1998 IEEE International Conference on Acoustics, Speech and Signal Processing
Number	23
Country/Territory	United States
City	Seattle
Period	12/05/1998 → 15/05/1998

Bibliographical note

Copyright: 1998 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE

Access to Document

10.1109/ICASSP.1998.675465

RiisFinal published version, 530 KB

OpenUrl availability

Full text

Cite this

@inproceedings{30cecbe555714342a4dfe282842fa83c,

title = "Hidden neural networks: application to speech recognition",

abstract = "We evaluate the hidden neural network HMM/NN hybrid on two speech recognition benchmark tasks; (1) task independent isolated word recognition on the Phonebook database, and (2) recognition of broad phoneme classes in continuous speech from the TIMIT database. It is shown how hidden neural networks (HNNs) with much fewer parameters than conventional HMMs and other hybrids can obtain comparable performance, and for the broad class task it is illustrated how the HNN can be applied as a purely transition based system, where acoustic context dependent transition probabilities are estimated by neural networks",

author = "Riis, \{S{\o}ren Kamaric\}",

note = "Copyright: 1998 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE; 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 1998 ; Conference date: 12-05-1998 Through 15-05-1998",

year = "1998",

doi = "10.1109/ICASSP.1998.675465",

language = "English",

isbn = "0-7803-4428-6",

volume = "2",

pages = "1117--1120",

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

publisher = "IEEE",

address = "United States",

}

Riis, SK 1998, Hidden neural networks: application to speech recognition. in Acoustics, Speech and Signal Processing, 1998. Proceedings of the 1998 IEEE International Conference on. vol. 2, IEEE, pp. 1117-1120, 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, Seattle, Washington, United States, 12/05/1998. https://doi.org/10.1109/ICASSP.1998.675465

TY - GEN

T1 - Hidden neural networks: application to speech recognition

AU - Riis, Søren Kamaric

N1 - Conference code: 23

PY - 1998

Y1 - 1998

N2 - We evaluate the hidden neural network HMM/NN hybrid on two speech recognition benchmark tasks; (1) task independent isolated word recognition on the Phonebook database, and (2) recognition of broad phoneme classes in continuous speech from the TIMIT database. It is shown how hidden neural networks (HNNs) with much fewer parameters than conventional HMMs and other hybrids can obtain comparable performance, and for the broad class task it is illustrated how the HNN can be applied as a purely transition based system, where acoustic context dependent transition probabilities are estimated by neural networks

AB - We evaluate the hidden neural network HMM/NN hybrid on two speech recognition benchmark tasks; (1) task independent isolated word recognition on the Phonebook database, and (2) recognition of broad phoneme classes in continuous speech from the TIMIT database. It is shown how hidden neural networks (HNNs) with much fewer parameters than conventional HMMs and other hybrids can obtain comparable performance, and for the broad class task it is illustrated how the HNN can be applied as a purely transition based system, where acoustic context dependent transition probabilities are estimated by neural networks

U2 - 10.1109/ICASSP.1998.675465

DO - 10.1109/ICASSP.1998.675465

M3 - Article in proceedings

SN - 0-7803-4428-6

VL - 2

SP - 1117

EP - 1120

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

PB - IEEE

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

Y2 - 12 May 1998 through 15 May 1998

ER -

Hidden neural networks: application to speech recognition

Abstract

Conference

Bibliographical note

Access to Document

OpenUrl availability

Fingerprint

Cite this