Title :
On recognition of spoken Bengali numerals
Author :
Ghanty, Sumit Kumar ; Shaikh, Soharab Hossain ; Chaki, Nabendu
Author_Institution :
A.K Choudhury Sch. of Inf. Technol., Univ. of Calcutta, Kolkata, India
Abstract :
This paper presents a method for recognizing isolated spoken Bengali numerals. Noisy audio samples have been considered as input in this study. Mel frequency cepstral coefficients (MFCC) have been used for extraction of feature from the audio samples. Vector quantization is applied to reduce the dimension of the feature vectors and to generate a vector codebook for the numerals. The classification is based on the dynamic time warping (DTW) and a minimum distance classifier based on Euclidean distance measure. Both the speaker dependent and speaker independent situations have been considered for checking accuracy. Results show the limitations of MFCC based standard speech processing approach in speaker independent spoken digit recognition scenario in the presence of noise.
Keywords :
cepstral analysis; feature extraction; natural language processing; speaker recognition; speech coding; time warp simulation; vector quantisation; Euclidean distance; Mel frequency cepstral coefficient; distance classifier; dynamic time warping; feature extraction; noisy audio sample; speaker Recognition; speech processing; spoken Bengali numeral; vector codebook; vector quantization; Classification algorithms; Mel frequency cepstral coefficient; Noise; Speech; Speech recognition; Support vector machine classification; Vector quantization; Euclidean distance classifier; Mel frequency cepstral coefficients; Spoken numeral recognition; dynamic time warping; vector quantization;
Conference_Titel :
Computer Information Systems and Industrial Management Applications (CISIM), 2010 International Conference on
Conference_Location :
Krackow
Print_ISBN :
978-1-4244-7817-0
DOI :
10.1109/CISIM.2010.5643692
