Phonetic-Attributes Dependent Speaker Verification, M.Sc. Thesis Sharif University of Technology ; Ghaemmaghami, Shahrokh (Supervisor)
Abstract
The purpose of this project is to improve current speaker verification techniques with short utterance using phonetic information extraction. I-vector technique is widely used in speaker verification systems. Different speakers span a subspace of universal acoustic space, which is usually modeled by “Universal Background model”. Speaker-specific subspace depends on the voice of speaker. In state-of-the-art speaker verification systems i-vectors are extracted by a factor analysis technique to represent speaker characteristics. Studies demonstrate that voiced phonemes contain more speaker-specific information than unvoiced. In this thesis we have classified voiced frames in order to exploit...
Cataloging briefPhonetic-Attributes Dependent Speaker Verification, M.Sc. Thesis Sharif University of Technology ; Ghaemmaghami, Shahrokh (Supervisor)
Abstract
The purpose of this project is to improve current speaker verification techniques with short utterance using phonetic information extraction. I-vector technique is widely used in speaker verification systems. Different speakers span a subspace of universal acoustic space, which is usually modeled by “Universal Background model”. Speaker-specific subspace depends on the voice of speaker. In state-of-the-art speaker verification systems i-vectors are extracted by a factor analysis technique to represent speaker characteristics. Studies demonstrate that voiced phonemes contain more speaker-specific information than unvoiced. In this thesis we have classified voiced frames in order to exploit...
Find in contentBookmark |
|