Neutral Speech to Target Speech Conversion by Prosodic Modification

The dynamics of prosodic features are utilized for speech emotion conversion. In particular, emotion conversion of neutral speech to sad, fear, anger and happy speech is accomplished. The prosodic features considered for the study are pitch contour and duration. Subjective listening test results show that the effectiveness of perception of emotion is better in the case of pitch contour and duration for the whole utterance. The results show that the converted sad, fear, angry speech are perceived very close to natural sad, fear, anger and happy emotions.