Title: Deep neural network-based phoneme classification of standard Khasi dialect in continuous speech
Authors: Bronson Syiem; L. Joyprakash Singh
Addresses: Department of Electronics and Communication Engineering, North Eastern Hill University, Shillong-793022, Meghalaya, India ' Department of Electronics and Communication Engineering, North Eastern Hill University, Shillong-793022, Meghalaya, India
Abstract: In this paper, deep neural network (DNN) is used to classify phonemes of the standard Khasi dialect which is one of the commonly used dialects in the state of Meghalaya. For this, clean speech data were recorded in the laboratory from native speakers. In the proposed system, a monophone and a triphone hidden Markov models (HMMs) were also built to compare the results obtained. Our finding showed that DNN outperformed the classification over the other two models with a classification accuracy of 89.70%.
Keywords: acoustic model; deep neural network; DNN; Gaussian mixture model; GMM; hidden Markov model; HMM; language model; Mel frequency cepstral coefficient; MFCC; phone error rate; PER; voice activity detection; VAD; word error rate; WER.
DOI: 10.1504/IJAPR.2019.104288
International Journal of Applied Pattern Recognition, 2019 Vol.6 No.1, pp.43 - 51
Received: 15 Oct 2018
Accepted: 09 Apr 2019
Published online: 02 Jan 2020 *