Supervector Extraction for Encoding Speaker and Phrase Information with Neural Networks for Text-Dependent Speaker Verification

Supervector Extraction for Encoding Speaker and Phrase Information with Neural Networks for Text-Dependent Speaker Verification

In this paper, we propose a new differentiable neural network with an alignment mechanism for text-dependent speaker verification. Unlike previous works, we do not extract the embedding of an utterance from the global average pooling of the temporal dimension. Our system replaces this reduction mech...

Full description

Bibliographic Details
Main Authors: Victoria Mingote, Antonio Miguel, Alfonso Ortega, Eduardo Lleida
Format: Article
Language:English
Published: MDPI AG 2019-08-01
Series:Applied Sciences
Subjects:
text-dependent speaker verification
HMM alignment
deep neural networks
supervectors
Online Access:https://www.mdpi.com/2076-3417/9/16/3295

Internet

https://www.mdpi.com/2076-3417/9/16/3295

Similar Items