Phoneme modeling for speech recognition in Kannada using Hidden Markov Model

Prashanth Kannadaguli, Ananthakrishna Thalengala

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

2 Citations (Scopus)

Abstract

We build an automatic phoneme recognition system based on Hidden Markov Modeling (HMM) which is a Dynamic modeling scheme. 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[1,2], we have used the same in speech feature extraction. Finally performance analysis of models in terms of Phoneme Error Rate (PER) justifies the fact that Dynamic modeling yields good results and can be used in developing Automatic Speech Recognition systems.

Original languageEnglish
Title of host publication2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems, SPICES 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781479918232
DOIs
Publication statusPublished - 21-04-2015
Event2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems, SPICES 2015 - Calicut, India
Duration: 19-02-201521-02-2015

Conference

Conference2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems, SPICES 2015
CountryIndia
CityCalicut
Period19-02-1521-02-15

Fingerprint

Hidden Markov models
Speech recognition
Acoustics
Speech analysis
Pattern recognition
Feature extraction

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Signal Processing
  • Software

Cite this

Kannadaguli, P., & Thalengala, A. (2015). Phoneme modeling for speech recognition in Kannada using Hidden Markov Model. In 2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems, SPICES 2015 [7091382] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SPICES.2015.7091382
Kannadaguli, Prashanth ; Thalengala, Ananthakrishna. / Phoneme modeling for speech recognition in Kannada using Hidden Markov Model. 2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems, SPICES 2015. Institute of Electrical and Electronics Engineers Inc., 2015.
@inproceedings{8d6064caf149430e9eedd12056ba4ef6,
title = "Phoneme modeling for speech recognition in Kannada using Hidden Markov Model",
abstract = "We build an automatic phoneme recognition system based on Hidden Markov Modeling (HMM) which is a Dynamic modeling scheme. 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[1,2], we have used the same in speech feature extraction. Finally performance analysis of models in terms of Phoneme Error Rate (PER) justifies the fact that Dynamic modeling yields good results and can be used in developing Automatic Speech Recognition systems.",
author = "Prashanth Kannadaguli and Ananthakrishna Thalengala",
year = "2015",
month = "4",
day = "21",
doi = "10.1109/SPICES.2015.7091382",
language = "English",
booktitle = "2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems, SPICES 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

Kannadaguli, P & Thalengala, A 2015, Phoneme modeling for speech recognition in Kannada using Hidden Markov Model. in 2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems, SPICES 2015., 7091382, Institute of Electrical and Electronics Engineers Inc., 2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems, SPICES 2015, Calicut, India, 19-02-15. https://doi.org/10.1109/SPICES.2015.7091382

Phoneme modeling for speech recognition in Kannada using Hidden Markov Model. / Kannadaguli, Prashanth; Thalengala, Ananthakrishna.

2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems, SPICES 2015. Institute of Electrical and Electronics Engineers Inc., 2015. 7091382.

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

TY - GEN

T1 - Phoneme modeling for speech recognition in Kannada using Hidden Markov Model

AU - Kannadaguli, Prashanth

AU - Thalengala, Ananthakrishna

PY - 2015/4/21

Y1 - 2015/4/21

N2 - We build an automatic phoneme recognition system based on Hidden Markov Modeling (HMM) which is a Dynamic modeling scheme. 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[1,2], we have used the same in speech feature extraction. Finally performance analysis of models in terms of Phoneme Error Rate (PER) justifies the fact that Dynamic modeling yields good results and can be used in developing Automatic Speech Recognition systems.

AB - We build an automatic phoneme recognition system based on Hidden Markov Modeling (HMM) which is a Dynamic modeling scheme. 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[1,2], we have used the same in speech feature extraction. Finally performance analysis of models in terms of Phoneme Error Rate (PER) justifies the fact that Dynamic modeling yields good results and can be used in developing Automatic Speech Recognition systems.

UR - http://www.scopus.com/inward/record.url?scp=84937132865&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84937132865&partnerID=8YFLogxK

U2 - 10.1109/SPICES.2015.7091382

DO - 10.1109/SPICES.2015.7091382

M3 - Conference contribution

BT - 2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems, SPICES 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Kannadaguli P, Thalengala A. Phoneme modeling for speech recognition in Kannada using Hidden Markov Model. In 2015 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems, SPICES 2015. Institute of Electrical and Electronics Engineers Inc. 2015. 7091382 https://doi.org/10.1109/SPICES.2015.7091382