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 Bronson Syiem;Sushanta Kabir Dutta;Juwesh Binong;Lairenlakpam Joyprakash Singh

作者机构：Department of Electronics and Communication EngineeringNorth-Eastern Hill UniversityShillong 793022 

出 版 物：《Journal of Electronic Science and Technology》 (电子科技学刊（英文版）)

年 卷 期：2021年第19卷第2期

页      面：155-162页

核心收录：

学科分类：0711[理学-系统科学] 07[理学] 

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 

主　　题：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) 

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 *** 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 *** each spectral feature,different hidden Markov model(HMM)based recognizers with variations in HMM states and different Gaussian mixture models(GMMs)were *** 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.