Backend of Voice based User Authentication System. Trained Voice Recognition using Mel-Frequency Cepstral Coefficients (MFCC) and Linear Prediction Coefficients (LPC).

The most popular feature matching algorithms for Voice Recognition are Dynamic Time Warping (DTW), Hidden Markov Model (HMM) and Vector Quantization (VQ). Here, I have used Vector Quantization.

As I gone through different accuracies in several voice sampples.It is possible to conclude that the program had its performance achieving as much as 70% of accuracy. But with large training data will increase the accuracy. As I thought to implement a noise cancellation system to minimize the effects of the environment in the result, but I could not find any reference regarding it ,and also the lack of knowledge in the probability theory that was necessary to implement this feature. And also I don't have that much of time to concentrate on this noise cancellation feature.In Future noice cancellation module will be trained.