Performance Analysis of Feature sets in Speaker Diarization techniques

Abstract

Speech is the most important communication among humans. Processing of speech signal has many strategies including speech coding, speaker recognition, speaker verification, etc. Speaker diarization is the pre-processing stage for many applications of speaker recognition systems. Speaker Diarization is the mission of determining “who Spoke when” for any audio recording that carries an unknown quantity of records and an unknown variety of audio systems. Speaker diarization has come to be achief era for many tasks like navigation, retrieval, or higher-level interference on audio data. It mainly performs three operations feature extraction, voice activity detection, and classification. In this paper, we’ve reviewed the few speaker diarization Techniques. The trendy speaker diarization structures finished nice outcomes. In this paper, few speaker diarization device performances are evaluated for Diarization mistakes, Tracking time, and False alarm.