Emotion recognition using LP residual at sub-segmental, segmental and supra-segmental levels

摘要

This paper is concerned with speech signal based emotion recognition. Linear Prediction (LP) residual mainly contains source specific emotional information. LP residual is derived by inverse filtering of the speech signal. For characterizing the basic emotions, LP residual has been explored at sub-segmental level, segmental level, supra-segmental level, respectively. Gaussian mixture models (GMMs) have been used as classifier. IIT Kharagpur Simulated Emotion Speech Corpus (IITKGP-SESC) and Berlin emotional database (Berlin-EMO-DB) are used as a database for this purpose. Average emotion recognition rate is observed to be 58.4%, 65.6% and 48% at sub-segmental level, segmental level and supra-segmental level, respectively.