This paper is concerned with the development of Back-propagation NeuralNetwork for Bangla Speech Recognition. In this paper, ten bangla digits wererecorded from ten speakers and have been recognized. The features of thesespeech digits were extracted by the method of Mel Frequency CepstralCoefficient (MFCC) analysis. The mfcc features of five speakers were used totrain the network with Back propagation algorithm. The mfcc features of tenbangla digit speeches, from 0 to 9, of another five speakers were used to testthe system. All the methods and algorithms used in this research wereimplemented using the features of Turbo C and C++ languages. From ourinvestigation it is seen that the developed system can successfully encode andanalyze the mfcc features of the speech signal to recognition. The developedsystem achieved recognition rate about 96.332% for known speakers (i.e.,speaker dependent) and 92% for unknown speakers (i.e., speaker independent).
展开▼
