Comparison of Khasi Speech Representations with Different Spectral Features and Hidden Markov States  

作　　者：Bronson Syiem Sushanta Kabir Dutta Juwesh Binong Lairenlakpam Joyprakash Singh 

机构地区：[1]Department of Electronics and Communication Engineering,North-Eastern Hill University,Shillong 793022

出　　处：《Journal of Electronic Science and Technology》2021年第2期155-162,共8页电子科技学刊（英文版）

基　　金：supported by the Visvesvaraya Ph.D.Scheme for Electronics and IT students launched by the Ministry of Electronics and Information Technology(MeiTY),Government of India under Grant No.PhD-MLA/4(95)/2015-2016.

摘　　要：In this paper,we present a comparison of Khasi speech representations with four different spectral features and novel extension towards the development of Khasi speech corpora.These four features include linear predictive coding(LPC),linear prediction cepstrum coefficient(LPCC),perceptual linear prediction(PLP),and Mel frequency cepstral coefficient(MFCC).The 10-hour speech data were used for training and 3-hour data for testing.For each spectral feature,different hidden Markov model(HMM)based recognizers with variations in HMM states and different Gaussian mixture models(GMMs)were built.The performance was evaluated by using the word error rate(WER).The experimental results show that MFCC provides a better representation for Khasi speech compared with the other three spectral features.

关 键 词：Acoustic model(AM) Gaussian mixture model(GMM) hidden Markov model(HMM) language model(LM) linear predictive coding(LPC) linear prediction cepstral coefficient(LPCC) Mel frequency cepstral coefficient(MFCC) perceptual linear prediction(PLP) 

分 类 号：TN912.3[电子电信—通信与信息系统]