Paper Title
Speech Emotion Recognition Using Iterative Clustering Technique

This paper proposes a method to recognize the emotion present in the speech signal using Iterative clustering technique. We propose Mel Frequency Perceptual Linear Predictive Cepstrum (MFPLPC) as a feature for recognizing the emotions. This feature is extracted from the speech and the clustering models are generated for each emotion. For the Speaker Independent classification technique, pre- processing is done on test speeches and features are extracted. In K- means clustering algorithm, the classification is based on the minimum distance between each test vector and centroid of clusters. Mean of the minimum distances for each speech is found out. The test speech belongs to the model which has minimum of averages corresponds to particular emotion. We obtained better recognition rate by using MFPLPC feature for a cluster size 1024. The results are obtained for SAVEE database using data of seven emotional states such as anger, disgust, fear, happy, sad, surprise and neutral. KeyWords- Emotion recognition (ER), Mel Frequency Perceptual Linear Predictive Cepstrum (MFPLPC), Vector Quantization (VQ).