A Novel Method for Speech Segmentation Based on Speaker's Characteristics

Speech Segmentation is the process change point detection for partitioning an input audio stream intoregions each of which corresponds to only one audio source or one speaker. One application of this systemis in Speaker Diarization systems. There are several methods for speaker segmentation; however, most ofthe Speaker Diarization Systems use BIC-based Segmentation methods. The main goal of this paper is topropose a new method for speaker segmentation with higher speed than the current methods - e.g. BIC -and acceptable accuracy. Our proposed method is based on the pitch frequency of the speech. Theaccuracy of this method is similar to the accuracy of common speaker segmentation methods. However, itscomputation cost is much less than theirs. We show that our method is about 2.4 times faster than the BICbasedmethod, while the average accuracy of pitch-based method is slightly higher than that of the BICbasedmethod.