Robust Features for Noisy Text-independent Speaker Identification Using GFCC Algorithm Combined to VAD and CMN Techniques

Provided by: Journal of Theoretical and Applied Information Technology
Topic: Enterprise Software
Format: PDF
A major problem of most speaker identification systems is their unsatisfactory robustness in noisy environments. The performance of automatic speaker identification systems degrade drastically in the presence of noise and other distortions, especially when there is a noise level mismatch between the training and testing environments. In this paper, the authors have studied a recently robust front-end algorithm based on Gammatone Frequency Cepstral Coefficients (GFCCs) associated to Voice Activity Detector (VAD) and Cepstral Mean Normalization (CMN) techniques.

Find By Topic