ISSN : 2349-6657

Text-Independent Speaker Verification– A Review

M. Selin and Dr.K. Preetha Mathew

Speech is the most natural means of communication among people. Speaker recognition is the process of identification of a speaker from his/her speech sample. This includes two methods: Speaker identification and Speaker verification. Speaker identification is the process of determining an unknown speaker’s identity, and speaker verification is the process of accepting or rejecting the identity claimed by a speaker. Speaker verification can be further classified into Text-dependent speaker verification (TDSV) and Text-independent speaker verification (TISV). Text-independent speaker verification continues to be one of the most active research areas in speech processing which aims to find whether an utterance pronounced by a particular speaker is similar to his/her pre-recorded utterance without limiting contents. A speaker verification system involves two main phases: the Training phase includes the development and enrollment of target speakers and the Testing phase for the evaluation of the identity of the speaker. From a training point of view, speaker models can be classified into generative and discriminative. Generative models like the Gaussian Mixture Model (GMM), estimate the feature distribution within each speaker. Discriminative models such as Support Vector Machine (SVM) and Deep Neural Network (DNN) in contrast, model the boundary between speakers. This paper gives a review of Text-Independent speaker verification using various techniques. The comparative study of various approaches for speaker verification, and the fundamentals like feature selection, feature modelling, and classification using different techniques were also discussed.

Speaker Verification, TISV, Feature Extraction, Gaussian Mixture Model (GMM), Deep Neural Network (DNN), CNN, Support Vector Machine (SVM), PLDA




Paper Submission Last Date

Notification of Acceptance

Camera Ready Paper Submission & Author's Registration

Date of Conference