Text-Independent Speaker Identification: A Survey

Abstract

Nowadays it is obvious that speakers can be identified from their voices. In this work the details of speaker identification from the real-time system point of view are discussed. The speaker identification systems can be subdivided into text-dependent and text-independent methods. Text-dependent systems require the speaker to utter a specific phrase (pin-code, password etc.), while a text-independent method should catch the characteristics of the speech irrespective of the text spoken. The system developed in this work is the latter, text-independent, meaning the system can identify the speaker regardless of what is being said.  This paper presents text-independent speaker identification system, which consists of mapping a speech signal from an unknown speaker to a database of known speakers, i.e. the system hasbeen trained with a number of speakers which the system can recognize. Speaker identification has been done with hidden Markova model and vector quantization.