Please use this identifier to cite or link to this item: https://ir.swu.ac.th/jspui/handle/123456789/12756
ชื่อเรื่อง: Comparison of feature extraction for accent dependent Thai speech recognition system
ผู้แต่ง: Tantisatirapong S.
Prasoproek C.
Phothisonothai M.
Keywords: Electrostatic devices
Extraction
Power spectral density
Radial basis function networks
Spectral density
Spectrographs
Speech recognition
Support vector machines
Energy spectral density
Feature extraction methods
Mel frequency cepstral co-efficient
Mel frequency cepstral coefficients (MFCC)
Power spectral densities (PSD)
Radial basis function kernels
Spectrograms
Speech recognition systems
Feature extraction
วันที่เผยแพร่: 2018
บทคัดย่อ: This paper aims to compare the feature extraction methods for accent dependent Thai speech from three regions including central, southern and northeastern regions. We investigate four frequency analysis methods: i.e., Energy Spectral Density (ESD), Power Spectral Density (PSD), Mel-Frequency Cepstral Coefficients (MFCC) and Spectrogram (SPT). Radial basis function kernel based on support vector machine is used as a classifier with 5-fold cross validation. The isolated speech data sets are recorded from 30 male and 30 female participants speaking the 10 Thai digits from 0 to 9. The MFCC-based feature gives better accuracy than ESD, PSD and SPT respectively. For within the same region, the MFCC-based feature provides average accuracy of 94.9% and 99.1% for male and female voices respectively. For the three regions, the MFCC-based feature provides average accuracy of 89.34% and 93.81% for male and female voices, respectively. © 2018 IEEE.
URI: https://ir.swu.ac.th/jspui/handle/123456789/12756
https://www.scopus.com/inward/record.uri?eid=2-s2.0-85057580394&doi=10.1109%2fCCE.2018.8465705&partnerID=40&md5=3527e08420dc09bae02d46e65b952d9c
Appears in Collections:Scopus 1983-2021

Files in This Item:
There are no files associated with this item.


Items in SWU repository are protected by copyright, with all rights reserved, unless otherwise indicated.