A comparison of Gaussian Mixture Modeling (GMM) and Hidden Markov Modeling (HMM) based approaches for Automatic Phoneme Recognition in Kannada

Prashanth Kannadaguli, Vidya Bhat

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

6 Citations (Scopus)

Abstract

We build and compare phoneme recognition systems based on Gaussian Mixture Modeling (GMM) which is a static modeling scheme and Hidden Markov Modeling (HMM) which is a Dynamic modeling scheme. Both models were built by using Stochastic pattern recognition and Acoustic phonetic schemes to recognise phonemes. Since our native language is Kannada, a rich South Indian Language, we have used 15 Kannada phonemes to train and test these models. Since Mel-Frequency Cepstral Coefficients (MFCC) are well known Acoustic features of speech, we have used the same in speech feature extraction. Finally performance analysis of both models in terms of Phoneme Error Rate (PER) justifies the fact that Dynamic modeling yields better results over Static modeling and can be used in developing Automatic Speech Recognition systems.

Original languageEnglish
Title of host publication2015 International Conference on Signal Processing and Communication, ICSC 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages257-260
Number of pages4
ISBN (Electronic)9781479967612
DOIs
Publication statusPublished - 01-01-2015
Externally publishedYes
Event2015 International Conference on Signal Processing and Communication, ICSC 2015 - Noida, India
Duration: 16-03-201518-03-2015

Conference

Conference2015 International Conference on Signal Processing and Communication, ICSC 2015
CountryIndia
CityNoida
Period16-03-1518-03-15

Fingerprint

Acoustics
Speech analysis
Speech recognition
Pattern recognition
Feature extraction

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Signal Processing

Cite this

Kannadaguli, P., & Bhat, V. (2015). A comparison of Gaussian Mixture Modeling (GMM) and Hidden Markov Modeling (HMM) based approaches for Automatic Phoneme Recognition in Kannada. In 2015 International Conference on Signal Processing and Communication, ICSC 2015 (pp. 257-260). [7150658] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICSPCom.2015.7150658
Kannadaguli, Prashanth ; Bhat, Vidya. / A comparison of Gaussian Mixture Modeling (GMM) and Hidden Markov Modeling (HMM) based approaches for Automatic Phoneme Recognition in Kannada. 2015 International Conference on Signal Processing and Communication, ICSC 2015. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 257-260
@inproceedings{fb2a3294d4304bceb54d414fb5935682,
title = "A comparison of Gaussian Mixture Modeling (GMM) and Hidden Markov Modeling (HMM) based approaches for Automatic Phoneme Recognition in Kannada",
abstract = "We build and compare phoneme recognition systems based on Gaussian Mixture Modeling (GMM) which is a static modeling scheme and Hidden Markov Modeling (HMM) which is a Dynamic modeling scheme. Both models were built by using Stochastic pattern recognition and Acoustic phonetic schemes to recognise phonemes. Since our native language is Kannada, a rich South Indian Language, we have used 15 Kannada phonemes to train and test these models. Since Mel-Frequency Cepstral Coefficients (MFCC) are well known Acoustic features of speech, we have used the same in speech feature extraction. Finally performance analysis of both models in terms of Phoneme Error Rate (PER) justifies the fact that Dynamic modeling yields better results over Static modeling and can be used in developing Automatic Speech Recognition systems.",
author = "Prashanth Kannadaguli and Vidya Bhat",
year = "2015",
month = "1",
day = "1",
doi = "10.1109/ICSPCom.2015.7150658",
language = "English",
pages = "257--260",
booktitle = "2015 International Conference on Signal Processing and Communication, ICSC 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

Kannadaguli, P & Bhat, V 2015, A comparison of Gaussian Mixture Modeling (GMM) and Hidden Markov Modeling (HMM) based approaches for Automatic Phoneme Recognition in Kannada. in 2015 International Conference on Signal Processing and Communication, ICSC 2015., 7150658, Institute of Electrical and Electronics Engineers Inc., pp. 257-260, 2015 International Conference on Signal Processing and Communication, ICSC 2015, Noida, India, 16-03-15. https://doi.org/10.1109/ICSPCom.2015.7150658

A comparison of Gaussian Mixture Modeling (GMM) and Hidden Markov Modeling (HMM) based approaches for Automatic Phoneme Recognition in Kannada. / Kannadaguli, Prashanth; Bhat, Vidya.

2015 International Conference on Signal Processing and Communication, ICSC 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 257-260 7150658.

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

TY - GEN

T1 - A comparison of Gaussian Mixture Modeling (GMM) and Hidden Markov Modeling (HMM) based approaches for Automatic Phoneme Recognition in Kannada

AU - Kannadaguli, Prashanth

AU - Bhat, Vidya

PY - 2015/1/1

Y1 - 2015/1/1

N2 - We build and compare phoneme recognition systems based on Gaussian Mixture Modeling (GMM) which is a static modeling scheme and Hidden Markov Modeling (HMM) which is a Dynamic modeling scheme. Both models were built by using Stochastic pattern recognition and Acoustic phonetic schemes to recognise phonemes. Since our native language is Kannada, a rich South Indian Language, we have used 15 Kannada phonemes to train and test these models. Since Mel-Frequency Cepstral Coefficients (MFCC) are well known Acoustic features of speech, we have used the same in speech feature extraction. Finally performance analysis of both models in terms of Phoneme Error Rate (PER) justifies the fact that Dynamic modeling yields better results over Static modeling and can be used in developing Automatic Speech Recognition systems.

AB - We build and compare phoneme recognition systems based on Gaussian Mixture Modeling (GMM) which is a static modeling scheme and Hidden Markov Modeling (HMM) which is a Dynamic modeling scheme. Both models were built by using Stochastic pattern recognition and Acoustic phonetic schemes to recognise phonemes. Since our native language is Kannada, a rich South Indian Language, we have used 15 Kannada phonemes to train and test these models. Since Mel-Frequency Cepstral Coefficients (MFCC) are well known Acoustic features of speech, we have used the same in speech feature extraction. Finally performance analysis of both models in terms of Phoneme Error Rate (PER) justifies the fact that Dynamic modeling yields better results over Static modeling and can be used in developing Automatic Speech Recognition systems.

UR - http://www.scopus.com/inward/record.url?scp=84943621902&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84943621902&partnerID=8YFLogxK

U2 - 10.1109/ICSPCom.2015.7150658

DO - 10.1109/ICSPCom.2015.7150658

M3 - Conference contribution

AN - SCOPUS:84943621902

SP - 257

EP - 260

BT - 2015 International Conference on Signal Processing and Communication, ICSC 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Kannadaguli P, Bhat V. A comparison of Gaussian Mixture Modeling (GMM) and Hidden Markov Modeling (HMM) based approaches for Automatic Phoneme Recognition in Kannada. In 2015 International Conference on Signal Processing and Communication, ICSC 2015. Institute of Electrical and Electronics Engineers Inc. 2015. p. 257-260. 7150658 https://doi.org/10.1109/ICSPCom.2015.7150658