Speaker change detection - an comparative study using support vector machines

Speaker change detection is important for automatic segmentation of multi speaker speech data into homogeneous segments with each segment containing the data of one speaker only. Existing approaches for speaker change detection are based on the dissimilarities of the distribution of the data before and after a speaker change point. In this paper, we propose a classification based technique for speaker change detection. Patterns extracted from the data around the speaker change points are used as positive examples. Patterns extracted from the data between the speaker change points are used as negative examples. The positive and negative examples are used in training a Support Vector Machine (SVM) for speaker change detection. The trained SVM is used to scan the continuous speech signal of multispeaker data and hypothesis the points of speaker change. The extraction of fixed length patterns from speaker are given as input to the Support Vector Machine. SVMs are used to classify the speaker change points and speaker no change points using speaker features. The performance of the system is evaluated for two speaker conversations. The dataset includes three conversation for each male-male, male-female, and female-female speaker conversations.