Real Time Speaker Recognition System for Hindi Words

Geeta Nijhawan, M.K Soni
2014 International Journal of Information Engineering and Electronic Business  
Real time speaker recognition is needed for various voice controlled applications. Background noise influences the overall efficiency of speaker recognition system and is still considered a challenge in Speaker Recognition System (SRS). In this paper MFCC feature is used along with VQLBG algorithm for designing SRS. A new approach for designing a Voice Activity Detector (VAD) has been proposed which can discriminate between silence and voice activity and this can significantly improve the
more » ... y improve the performance of SRS under noisy conditions. MFCC feature is extracted from the input speech and then vector quantization of the extracted MFCC features is done using VQLBG algorithm. Speaker identification is done by comparing the features of a newly recorded voice with the database under a specific threshold using Euclidean distance approach. The entire processing is done using MATLAB tool.The experimental result shows that the proposed method gives good results.
doi:10.5815/ijieeb.2014.02.04 fatcat:qrddfpamkvclxfiwcy37woe7mq