Voice Based Shapes Recognition using Mel Frequency Cepstrum Coefficents
Publication Date : 30/03/2016
Signal processing is computing an extracting the feature set is an important stage in any speech recognition system. The feasible feature set is still not yet decided though the vast efforts of researchers. There are many types of features, which are derived distinct and have good impact on the recognition rate. This project presents one of the techniques to take out the feature set from a speech signal, which can be used in speech recognition systems. The key is to adjust the speech wave form to parametric representation. To achieve this, we have first made a comparative study of the Mel Frequency Cepstrum Coefficients approach. The voice based biometric system is based on isolated or single word recognition. A particular microphone pronounce the words once in the training session so as to train and store the accentuate of the way in word. Later in the testing session the user pronounce the word again in order to achieve recognition if there is a same content. The quality vectors unique to that speaker are obtained in the training phase and this is made use of later on enter the parameters to the same microphone who once again pronounce the same word in the testing phase. At this stage with parameter value can also test the system.
No. of Downloads :