No Detail Found!

Combining Features at Vector Level for Higher Speed and Accuracy of Speaker Identification.

Author(s)
M Afzal Department of Computer Science and Engineering, University of Engineering and Technology Lahore, Pakistan
T Ahmad Department of Computer Science and Engineering, University of Engineering and Technology Lahore, Pakistan
MF Hayat Department of Computer Science and Engineering, University of Engineering and Technology Lahore, Pakistan
HMS Asif Department of Computer Science and Engineering, University of Engineering and Technology Lahore, Pakistan
KH Asif Department of Computer Science and Engineering, University of Engineering and Technology Lahore, Pakistan
Abstract
The study on combination of different feature vectors for the accuracy of Speaker Identification (SI) based on Vector Quantization (VQ) performed well when compared with other paradigms. Feature vectors based on Mel-frequency Cepstral Coefficients (MFCC) and Linear Predictive Codes (LPC) were combined. Texas Instrument and Massachusetts Institute of Technology (TIMIT) database containing 630 speakers were registered and tested for feature level combination studies. LPC feature vector manifested lower accuracy than MFCC when used as such. The combination of feature vectors through sum (MFCC+LPC) and difference (MFCC-LPC) were studied. For 42% of cases of codebook sizes studied (MFCC+LPC) gave higher accuracy than simple MFCC. The accuracy of (MFCC-LPC) combination results was better than simple MFCC for 93% of cases of codebook sizes studied. The accuracy enhancement could be used to reduce the time of speaker identification to half by using half sized codebooks of (MFCC-LPC)feature vectors with same or higher accuracy as compared to codebooks of simple MFCC feature vectors.
Publication Details
Page(s) 59-63
DOI DOI not available
Published Journal: Pakistan Journal of Science, Volume: 67, Issue: 1, Year: 2015

Request the full text from PASTIC

References
Afzal , M. and S.,Haq, 2010.Accelerating Vector quantization based speaker identification, 6 1046 -1050
Afzal , M.,M.Maud,A.Akbar, 2012.Toggling and circular partial distortion elimination algorithms to speedup speaker identification based on vector quantization,Pakistan J. Engin. and Appl. Sci., 11 1 -13
Aida , K.,C.Ardil,S.Rustamov, 2006.Investigation of combined use of MFCC and LPC features in speech recognition systems,World Academy of Science 2 74 -80
Bei , H.,andR. Gray, 1985.An improvement of the minimum distortion encoding algorithm for vector quantization,IEEE Transactions on Communication 33 1132 -1133
Garofolo , J.,L.Lamel,J.Fiscus,D.Pallett,D.Dahlgren,V.Zue,DARPA TIMIT, 1993.Acoustic phonetic continuous speech corpus, (NISTIR Publication No, 4930 -
Glaeser , A.,F.Bimbot,NSW , Australia, 1998.Steps towards the integration of speaker recognition in real world telecom applications,Proceeding of International Conference on Spoken Language Processing -
Kinnunen , T.,P.Franti, 2006.Real-time speaker identification and verification,IEEE Transactions on Audio and Language Processing 14 277 -288
Kinnunen , T.,T.Kilpelainen,P.Franti, 2000.Marbella,Proceeding of International Conference (IASTED) Signal Processing and Communications (SPC 222 -227
Kinnunen , T.,H.Li, 2010.An overview of text-independent speaker recognition: from features to supervectors, Speech Communication, 52 12 -40
Mashao , D.,M.Skosan, 2006.Combining classifier decisions for robust speaker identification,Pattern Recognition, Elsevier 1 147 -155
Quatieri , T.,Prentice HallPTR, 2002.Discrete-time speech signal processing principles and practice, -
0

Citations

0

Downloads

137

Views