Title :
Two-dimensional root cepstrum as feature extraction method for speech recognition
Author :
Chilton, E. ; Marvi, H.
Author_Institution :
Centre for Vision , Speech & Signal, Univ. of Surrey, Guildford, UK
fDate :
5/15/2003 12:00:00 AM
Abstract :
A novel and simple feature extraction method for speech recognition using the two-dimensional root cepstrum (TDRC) has been introduced. Because of the adjustable root index parameter (γ) it has some advantages over the original two-dimensional cepstrum (TDC). Experiments on isolated word recognition using the TIMIT data base show promising results compared with the original TDC.
Keywords :
cepstral analysis; discrete Fourier transforms; feature extraction; speech recognition; TIMIT data base; adjustable root index parameter; feature extraction method; inverse two-dimensional discrete Fourier transform; isolated word recognition; recognition accuracy; speech recognition; two-dimensional root cepstrum;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:20030522
