Speaker and speech recognition using pattern recognition approach in clean and noisy environment
International Journal of Latest Trends in Engineering and Technology · 2017
Speech signal consists of information regarding message, language, speaker and individual's emotions.Though speaker recognition and speech recognition are different fields but still both fields are overlapped.Combined speaker and speech recognition system is having different applications in human life from security related areas to aids for handicapped persons.This paper presents the speaker and speech recognition using pattern recognition approach.In this Mel Frequency Cepstral Coefficient (MFCC) are used as feature extraction method and for the classification, Dynamic Time Warping (DTW) algorithm is used.The recognition is done in clean as well as in noisy environment and accuracy is evaluated in terms of different parameters like false positive (FP), false negative (FN), true positive (TP) and true negative (TN).