Application of Teager Energy Operator on Linear and Mel Scales for Whispered Speech Recognition

This paper presents experimental results on whispered speech recognition based Teager Energy Operator for linear and mel cepstral coefficients including the Cepstral Mean Subtraction normalization technique. The feature vectors taken into consideration are Linear Frequency Coefficients, Mel Coefficients Coefficients. A speaker dependent scenario is used. For process, Dynamic Time Warping Hidden Markov Models methods applied. Results show a respectable improvement in as achieved by using with Subtraction.