Comparing NN paradigms in hybrid NN/HMM speech recognition using tied posteriors

J. Stadermann; G. Rigoll

doi:10.1109/ASRU.2003.1318409

Comparing NN paradigms in hybrid NN/HMM speech recognition using tied posteriors

J. Stadermann, G. Rigoll

Chair of Human-Machine Communication

Technical University of Munich

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

4 Scopus citations

Abstract

Hybrid NN/HMM acoustic modeling is nowadays an established alternative approach in automatic speech recognition technology. A comparison of feed-forward and recurrent neural network topologies integrated in the tied-posteriors framework is presented. We give some insights in the training process of the networks estimating class posterior probabilities and show how the net's quality can be determined by introducing a new measurement prior to evaluating the complete ASR system. Finally we demonstrate the flexibility of the tied-posteriors framework by showing results for different context independent and context dependent acoustic models all based on the same system structure.

Original language	English
Title of host publication	2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	89-93
Number of pages	5
ISBN (Electronic)	0780379802, 9780780379800
DOIs	https://doi.org/10.1109/ASRU.2003.1318409
State	Published - 2003
Event	IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003 - St. Thomas, United States Duration: 30 Nov 2003 → 4 Dec 2003

Publication series

Name	2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003

Conference

Conference	IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003
Country/Territory	United States
City	St. Thomas
Period	30/11/03 → 4/12/03

Access to Document

10.1109/ASRU.2003.1318409

Cite this

Stadermann, J., & Rigoll, G. (2003). Comparing NN paradigms in hybrid NN/HMM speech recognition using tied posteriors. In 2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003 (pp. 89-93). Article 1318409 (2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ASRU.2003.1318409

@inproceedings{3950a7588c6547e0bc84c3b5beed9916,

title = "Comparing NN paradigms in hybrid NN/HMM speech recognition using tied posteriors",

abstract = "Hybrid NN/HMM acoustic modeling is nowadays an established alternative approach in automatic speech recognition technology. A comparison of feed-forward and recurrent neural network topologies integrated in the tied-posteriors framework is presented. We give some insights in the training process of the networks estimating class posterior probabilities and show how the net's quality can be determined by introducing a new measurement prior to evaluating the complete ASR system. Finally we demonstrate the flexibility of the tied-posteriors framework by showing results for different context independent and context dependent acoustic models all based on the same system structure.",

author = "J. Stadermann and G. Rigoll",

note = "Publisher Copyright: {\textcopyright} 2003 IEEE.; IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003 ; Conference date: 30-11-2003 Through 04-12-2003",

year = "2003",

doi = "10.1109/ASRU.2003.1318409",

language = "English",

series = "2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "89--93",

booktitle = "2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003",

}

Stadermann, J & Rigoll, G 2003, Comparing NN paradigms in hybrid NN/HMM speech recognition using tied posteriors. in 2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003., 1318409, 2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003, Institute of Electrical and Electronics Engineers Inc., pp. 89-93, IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003, St. Thomas, United States, 30/11/03. https://doi.org/10.1109/ASRU.2003.1318409

Comparing NN paradigms in hybrid NN/HMM speech recognition using tied posteriors. / Stadermann, J.; Rigoll, G.
2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003. Institute of Electrical and Electronics Engineers Inc., 2003. p. 89-93 1318409 (2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003).

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

TY - GEN

T1 - Comparing NN paradigms in hybrid NN/HMM speech recognition using tied posteriors

AU - Stadermann, J.

AU - Rigoll, G.

PY - 2003

Y1 - 2003

N2 - Hybrid NN/HMM acoustic modeling is nowadays an established alternative approach in automatic speech recognition technology. A comparison of feed-forward and recurrent neural network topologies integrated in the tied-posteriors framework is presented. We give some insights in the training process of the networks estimating class posterior probabilities and show how the net's quality can be determined by introducing a new measurement prior to evaluating the complete ASR system. Finally we demonstrate the flexibility of the tied-posteriors framework by showing results for different context independent and context dependent acoustic models all based on the same system structure.

AB - Hybrid NN/HMM acoustic modeling is nowadays an established alternative approach in automatic speech recognition technology. A comparison of feed-forward and recurrent neural network topologies integrated in the tied-posteriors framework is presented. We give some insights in the training process of the networks estimating class posterior probabilities and show how the net's quality can be determined by introducing a new measurement prior to evaluating the complete ASR system. Finally we demonstrate the flexibility of the tied-posteriors framework by showing results for different context independent and context dependent acoustic models all based on the same system structure.

UR - http://www.scopus.com/inward/record.url?scp=33745188657&partnerID=8YFLogxK

U2 - 10.1109/ASRU.2003.1318409

DO - 10.1109/ASRU.2003.1318409

M3 - Conference contribution

AN - SCOPUS:33745188657

T3 - 2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003

SP - 89

EP - 93

BT - 2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003

Y2 - 30 November 2003 through 4 December 2003

ER -

Stadermann J, Rigoll G. Comparing NN paradigms in hybrid NN/HMM speech recognition using tied posteriors. In 2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003. Institute of Electrical and Electronics Engineers Inc. 2003. p. 89-93. 1318409. (2003 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2003). doi: 10.1109/ASRU.2003.1318409

Comparing NN paradigms in hybrid NN/HMM speech recognition using tied posteriors

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this