Generation of GMM Weights by Dirichlet Distribution and Model Selection Using Information Criterion for Malayalam Speech Recognition Lekshmi Krishna Ramachandran 1,2(B ) and Sherly Elizabeth 2 1 Bharathiar University, Coimbatore, Tamil Nadu, India lekshmi.kr@iiitmk.ac.in 2 Indian Institute of Information Technology and Management-Kerala, Thiruvananthapuram, India sherly@iiitmk.ac.in Abstract. Automatic Speech Recognition is a computer-driven tran- scription of spoken-language into human-readable text. This paper is focused on the development of an acoustic model for medium vocabu- lary, context independent, isolated Malayalam Speech Recognizer using Hidden Markov Model (HMM). In this work, the emission probabilities of syllables, based on HMMs are estimated from the Gaussian Mixture Model (GMM). Mel Frequency Cepstral Coefficient (MFCC) technique is used for feature extraction from the input speech. The generation of mixture weights for GMMs is done by implementing Dirichlet Dis- tribution. The efficiency of thus generated Gaussian Mixture Model is verified with different Information Criteria namely Akaike Information Criterion, Bayes Information Criterion, Corrected AIC, Kullback Linear Information Criterion, corrected KIC and Approximated KIC (KICc, AKICc). The accuracy of medium vocabulary, speaker dependent and isolated Malayalam speech corpus for a single Gaussian is 90.91% and Word Error Rate (WER) is 11.9%. The word accuracy and WER of the system are calculated based on the experiments conducted for multi- variate Gaussians. For Gaussian mixture five, a better word accuracy of 95.24% along with a WER of 4.76% is attained and the same is verified using Information Criteria. Keywords: Acoustic model · Akaike information criterion Bayes information criterion · HMM · GMM · Dirichlet distribution ASR · MFCC · Kullback information criterion Bias correction of Kullback information criterion Approximation of Kullback information criterion · Word error rate Kerala State Council of Science Technology and Environment-KSCSTE. L. Krishna Ramachandran—Research Scholar S. Elizabeth—Professor c  Springer Nature Switzerland AG 2018 U. S. Tiwary (Ed.): IHCI 2018, LNCS 11278, pp. 111–122, 2018. https://doi.org/10.1007/978-3-030-04021-5_11