Kannada word recognition system using HTK

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

1 Citation (Scopus)

Abstract

In the present work, speech recognition system for Kannada language has been implemented using the Hidden Markov Tool Kit (HTK). The system performance is comparatively studied and evaluated for syllable and phone level models. The Kannada word dictionary of size about 110 words is used in the study and Mel frequency cepstral coefficients (MFCC) are computed in acoustic front-end processing. The system is designed to recognize isolated utterances of Kannada words, which are recorded from a Kannada short story. Baum-Welch algorithm is used to train the Hidden Markov Model (HMM) and Viterbi algorithm for decoding process. The objective of this study is to compare the performances of phone-level and syllable-level acoustical models for small to medium sized Kannada language vocabulary. The results are part of the on-going research work on large vocabulary continuous speech recognition system for Kannada language. Average word recognition accuracy of 97.1% for syllable-level modeling and 98.6% for phone-level modeling has been reported. Analysis of system performance also carried out based on the confusion matrices.

Original languageEnglish
Title of host publication12th IEEE International Conference Electronics, Energy, Environment, Communication, Computer, Control
Subtitle of host publication(E3-C3), INDICON 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781467373999
DOIs
Publication statusPublished - 29-03-2016
Event12th IEEE International Conference Electronics, Energy, Environment, Communication, Computer, Control, INDICON 2015 - New Delhi, India
Duration: 17-12-201520-12-2015

Conference

Conference12th IEEE International Conference Electronics, Energy, Environment, Communication, Computer, Control, INDICON 2015
CountryIndia
CityNew Delhi
Period17-12-1520-12-15

Fingerprint

Continuous speech recognition
Viterbi algorithm
Hidden Markov models
Glossaries
Speech recognition
Decoding
Acoustics
Processing

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Hardware and Architecture
  • Energy Engineering and Power Technology
  • Renewable Energy, Sustainability and the Environment
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Ananthakrishna, T., Maithri, M., & Shama, K. (2016). Kannada word recognition system using HTK. In 12th IEEE International Conference Electronics, Energy, Environment, Communication, Computer, Control: (E3-C3), INDICON 2015 [7443122] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/INDICON.2015.7443122
Ananthakrishna, T. ; Maithri, M. ; Shama, Kumara. / Kannada word recognition system using HTK. 12th IEEE International Conference Electronics, Energy, Environment, Communication, Computer, Control: (E3-C3), INDICON 2015. Institute of Electrical and Electronics Engineers Inc., 2016.
@inproceedings{41504c1205934f908facb083d79a7c01,
title = "Kannada word recognition system using HTK",
abstract = "In the present work, speech recognition system for Kannada language has been implemented using the Hidden Markov Tool Kit (HTK). The system performance is comparatively studied and evaluated for syllable and phone level models. The Kannada word dictionary of size about 110 words is used in the study and Mel frequency cepstral coefficients (MFCC) are computed in acoustic front-end processing. The system is designed to recognize isolated utterances of Kannada words, which are recorded from a Kannada short story. Baum-Welch algorithm is used to train the Hidden Markov Model (HMM) and Viterbi algorithm for decoding process. The objective of this study is to compare the performances of phone-level and syllable-level acoustical models for small to medium sized Kannada language vocabulary. The results are part of the on-going research work on large vocabulary continuous speech recognition system for Kannada language. Average word recognition accuracy of 97.1{\%} for syllable-level modeling and 98.6{\%} for phone-level modeling has been reported. Analysis of system performance also carried out based on the confusion matrices.",
author = "T. Ananthakrishna and M. Maithri and Kumara Shama",
year = "2016",
month = "3",
day = "29",
doi = "10.1109/INDICON.2015.7443122",
language = "English",
booktitle = "12th IEEE International Conference Electronics, Energy, Environment, Communication, Computer, Control",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

Ananthakrishna, T, Maithri, M & Shama, K 2016, Kannada word recognition system using HTK. in 12th IEEE International Conference Electronics, Energy, Environment, Communication, Computer, Control: (E3-C3), INDICON 2015., 7443122, Institute of Electrical and Electronics Engineers Inc., 12th IEEE International Conference Electronics, Energy, Environment, Communication, Computer, Control, INDICON 2015, New Delhi, India, 17-12-15. https://doi.org/10.1109/INDICON.2015.7443122

Kannada word recognition system using HTK. / Ananthakrishna, T.; Maithri, M.; Shama, Kumara.

12th IEEE International Conference Electronics, Energy, Environment, Communication, Computer, Control: (E3-C3), INDICON 2015. Institute of Electrical and Electronics Engineers Inc., 2016. 7443122.

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

TY - GEN

T1 - Kannada word recognition system using HTK

AU - Ananthakrishna, T.

AU - Maithri, M.

AU - Shama, Kumara

PY - 2016/3/29

Y1 - 2016/3/29

N2 - In the present work, speech recognition system for Kannada language has been implemented using the Hidden Markov Tool Kit (HTK). The system performance is comparatively studied and evaluated for syllable and phone level models. The Kannada word dictionary of size about 110 words is used in the study and Mel frequency cepstral coefficients (MFCC) are computed in acoustic front-end processing. The system is designed to recognize isolated utterances of Kannada words, which are recorded from a Kannada short story. Baum-Welch algorithm is used to train the Hidden Markov Model (HMM) and Viterbi algorithm for decoding process. The objective of this study is to compare the performances of phone-level and syllable-level acoustical models for small to medium sized Kannada language vocabulary. The results are part of the on-going research work on large vocabulary continuous speech recognition system for Kannada language. Average word recognition accuracy of 97.1% for syllable-level modeling and 98.6% for phone-level modeling has been reported. Analysis of system performance also carried out based on the confusion matrices.

AB - In the present work, speech recognition system for Kannada language has been implemented using the Hidden Markov Tool Kit (HTK). The system performance is comparatively studied and evaluated for syllable and phone level models. The Kannada word dictionary of size about 110 words is used in the study and Mel frequency cepstral coefficients (MFCC) are computed in acoustic front-end processing. The system is designed to recognize isolated utterances of Kannada words, which are recorded from a Kannada short story. Baum-Welch algorithm is used to train the Hidden Markov Model (HMM) and Viterbi algorithm for decoding process. The objective of this study is to compare the performances of phone-level and syllable-level acoustical models for small to medium sized Kannada language vocabulary. The results are part of the on-going research work on large vocabulary continuous speech recognition system for Kannada language. Average word recognition accuracy of 97.1% for syllable-level modeling and 98.6% for phone-level modeling has been reported. Analysis of system performance also carried out based on the confusion matrices.

UR - http://www.scopus.com/inward/record.url?scp=84994285906&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84994285906&partnerID=8YFLogxK

U2 - 10.1109/INDICON.2015.7443122

DO - 10.1109/INDICON.2015.7443122

M3 - Conference contribution

BT - 12th IEEE International Conference Electronics, Energy, Environment, Communication, Computer, Control

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Ananthakrishna T, Maithri M, Shama K. Kannada word recognition system using HTK. In 12th IEEE International Conference Electronics, Energy, Environment, Communication, Computer, Control: (E3-C3), INDICON 2015. Institute of Electrical and Electronics Engineers Inc. 2016. 7443122 https://doi.org/10.1109/INDICON.2015.7443122