Phoneme modeling for speech recognition in Kannada using Hidden Markov Model

Author

Kannadaguli, Prashanth ; Thalengala, Ananthakrishna

Author_Institution

Dept. of Electron. & Commun. Eng., Manipal Inst. of Technol., Manipal, India

fYear

2015

fDate

19-21 Feb. 2015

Firstpage

1

Lastpage

5

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.

Keywords

acoustic signal processing; cepstral analysis; feature extraction; hidden Markov models; natural language processing; speech recognition; stochastic processes; Kannada; MFCC; PER; South Indian Language; acoustic phonetic schemes; automatic phoneme recognition system; dynamic modeling scheme; hidden Markov model; mel-frequency cepstral coefficients; phoneme error rate; phoneme modeling; speech feature extraction; stochastic pattern recognition; Hidden Markov models; Mel frequency cepstral coefficient; Speech; Speech recognition; Testing; Training; HMM; Kannada; MFCC; PER; Pattern Recognition; Phoneme Modeling;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing, Informatics, Communication and Energy Systems (SPICES), 2015 IEEE International Conference on

Conference_Location

Kozhikode

Type

conf

DOI

10.1109/SPICES.2015.7091382

Filename

7091382