Apply fuzzy vector quantization to improve the observation-based Discrete Hidden Markov Model An example on electroencephalogram (EEG) signal recognition

Shing Tai Pan, Sheng-Fu Liang, Tzung Pei Hong, Jian Hong Zeng

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

This paper applies fuzzy vector quantization (FVQ) to the modeling of observation-based Discrete Hidden Markov Model (DHMM) and then to improve the speech recognition rate for the Mandarin speech. Vector quantization based on a codebook is a fundamental process to recognize the speech signal by DHMM. A codebook will be first trained by K-means algorithms using Mandarin training speech. Then, based on the trained codebook, the speech features are quantized by the fuzzy sets defined on each vectors of the codebook. Subsequently, the quantized speech features are statistically applied to train the model of DHMM for the speech recognition. All the speech features to be recognized should go through the FVQ based on the fuzzy codebook before being fed into the DHMM model for recognition. Experimental results in this paper shows that the speech recognition rate can be improved by using FVQ algorithm to train the model of DHMM.

Original languageEnglish
Title of host publicationFUZZ 2011 - 2011 IEEE International Conference on Fuzzy Systems - Proceedings
Pages1674-1680
Number of pages7
DOIs
Publication statusPublished - 2011
Event2011 IEEE International Conference on Fuzzy Systems, FUZZ 2011 - Taipei, Taiwan
Duration: 2011 Jun 272011 Jun 30

Other

Other2011 IEEE International Conference on Fuzzy Systems, FUZZ 2011
CountryTaiwan
CityTaipei
Period11-06-2711-06-30

Fingerprint

Vector Quantization
Codebook
Vector quantization
Hidden Markov models
Electroencephalography
Markov Model
Speech Recognition
Speech recognition
K-means Algorithm
Speech Signal
Fuzzy Sets
Fuzzy sets
Observation
Electroencephalogram
Speech
Model
Experimental Results
Modeling

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Software
  • Artificial Intelligence
  • Applied Mathematics

Cite this

Pan, S. T., Liang, S-F., Hong, T. P., & Zeng, J. H. (2011). Apply fuzzy vector quantization to improve the observation-based Discrete Hidden Markov Model An example on electroencephalogram (EEG) signal recognition. In FUZZ 2011 - 2011 IEEE International Conference on Fuzzy Systems - Proceedings (pp. 1674-1680). [6007714] https://doi.org/10.1109/FUZZY.2011.6007714
Pan, Shing Tai ; Liang, Sheng-Fu ; Hong, Tzung Pei ; Zeng, Jian Hong. / Apply fuzzy vector quantization to improve the observation-based Discrete Hidden Markov Model An example on electroencephalogram (EEG) signal recognition. FUZZ 2011 - 2011 IEEE International Conference on Fuzzy Systems - Proceedings. 2011. pp. 1674-1680
@inproceedings{293d6c2330a6409cac29fea173c86ee2,
title = "Apply fuzzy vector quantization to improve the observation-based Discrete Hidden Markov Model An example on electroencephalogram (EEG) signal recognition",
abstract = "This paper applies fuzzy vector quantization (FVQ) to the modeling of observation-based Discrete Hidden Markov Model (DHMM) and then to improve the speech recognition rate for the Mandarin speech. Vector quantization based on a codebook is a fundamental process to recognize the speech signal by DHMM. A codebook will be first trained by K-means algorithms using Mandarin training speech. Then, based on the trained codebook, the speech features are quantized by the fuzzy sets defined on each vectors of the codebook. Subsequently, the quantized speech features are statistically applied to train the model of DHMM for the speech recognition. All the speech features to be recognized should go through the FVQ based on the fuzzy codebook before being fed into the DHMM model for recognition. Experimental results in this paper shows that the speech recognition rate can be improved by using FVQ algorithm to train the model of DHMM.",
author = "Pan, {Shing Tai} and Sheng-Fu Liang and Hong, {Tzung Pei} and Zeng, {Jian Hong}",
year = "2011",
doi = "10.1109/FUZZY.2011.6007714",
language = "English",
isbn = "9781424473175",
pages = "1674--1680",
booktitle = "FUZZ 2011 - 2011 IEEE International Conference on Fuzzy Systems - Proceedings",

}

Pan, ST, Liang, S-F, Hong, TP & Zeng, JH 2011, Apply fuzzy vector quantization to improve the observation-based Discrete Hidden Markov Model An example on electroencephalogram (EEG) signal recognition. in FUZZ 2011 - 2011 IEEE International Conference on Fuzzy Systems - Proceedings., 6007714, pp. 1674-1680, 2011 IEEE International Conference on Fuzzy Systems, FUZZ 2011, Taipei, Taiwan, 11-06-27. https://doi.org/10.1109/FUZZY.2011.6007714

Apply fuzzy vector quantization to improve the observation-based Discrete Hidden Markov Model An example on electroencephalogram (EEG) signal recognition. / Pan, Shing Tai; Liang, Sheng-Fu; Hong, Tzung Pei; Zeng, Jian Hong.

FUZZ 2011 - 2011 IEEE International Conference on Fuzzy Systems - Proceedings. 2011. p. 1674-1680 6007714.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Apply fuzzy vector quantization to improve the observation-based Discrete Hidden Markov Model An example on electroencephalogram (EEG) signal recognition

AU - Pan, Shing Tai

AU - Liang, Sheng-Fu

AU - Hong, Tzung Pei

AU - Zeng, Jian Hong

PY - 2011

Y1 - 2011

N2 - This paper applies fuzzy vector quantization (FVQ) to the modeling of observation-based Discrete Hidden Markov Model (DHMM) and then to improve the speech recognition rate for the Mandarin speech. Vector quantization based on a codebook is a fundamental process to recognize the speech signal by DHMM. A codebook will be first trained by K-means algorithms using Mandarin training speech. Then, based on the trained codebook, the speech features are quantized by the fuzzy sets defined on each vectors of the codebook. Subsequently, the quantized speech features are statistically applied to train the model of DHMM for the speech recognition. All the speech features to be recognized should go through the FVQ based on the fuzzy codebook before being fed into the DHMM model for recognition. Experimental results in this paper shows that the speech recognition rate can be improved by using FVQ algorithm to train the model of DHMM.

AB - This paper applies fuzzy vector quantization (FVQ) to the modeling of observation-based Discrete Hidden Markov Model (DHMM) and then to improve the speech recognition rate for the Mandarin speech. Vector quantization based on a codebook is a fundamental process to recognize the speech signal by DHMM. A codebook will be first trained by K-means algorithms using Mandarin training speech. Then, based on the trained codebook, the speech features are quantized by the fuzzy sets defined on each vectors of the codebook. Subsequently, the quantized speech features are statistically applied to train the model of DHMM for the speech recognition. All the speech features to be recognized should go through the FVQ based on the fuzzy codebook before being fed into the DHMM model for recognition. Experimental results in this paper shows that the speech recognition rate can be improved by using FVQ algorithm to train the model of DHMM.

UR - http://www.scopus.com/inward/record.url?scp=80053066747&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80053066747&partnerID=8YFLogxK

U2 - 10.1109/FUZZY.2011.6007714

DO - 10.1109/FUZZY.2011.6007714

M3 - Conference contribution

AN - SCOPUS:80053066747

SN - 9781424473175

SP - 1674

EP - 1680

BT - FUZZ 2011 - 2011 IEEE International Conference on Fuzzy Systems - Proceedings

ER -

Pan ST, Liang S-F, Hong TP, Zeng JH. Apply fuzzy vector quantization to improve the observation-based Discrete Hidden Markov Model An example on electroencephalogram (EEG) signal recognition. In FUZZ 2011 - 2011 IEEE International Conference on Fuzzy Systems - Proceedings. 2011. p. 1674-1680. 6007714 https://doi.org/10.1109/FUZZY.2011.6007714