Loading...
Loading...

Go to the content (press return)

DNN speaker embeddings using autoencoder pre-training

Author
Khan, U.; Hernando, J.
Type of activity
Presentation of work at congresses
Name of edition
27th European Signal Processing Conference
Date of publication
2019
Presentation's date
2019-09-03
Book of congress proceedings
27th EUSIPCO 2019 European Signal Processing Conference: A Coruña, Spain: September 2-6, 2019
First page
1
Last page
5
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
DOI
10.23919/EUSIPCO.2019.8902945
Project funding
Deep learning technologies for speech and audio processing
Repository
http://hdl.handle.net/2117/175406 Open in new window
URL
https://ieeexplore.ieee.org/document/8902945 Open in new window
Abstract
Over the last years, i-vectors have been the state-of-the-art approach in speaker recognition. Recent improvements in deep learning have increased the discriminative quality of i-vectors. However, deep learning architectures require a large amount of labeled background data which is difficult in practice. The aim of this paper is to propose an alternative scheme in order to reduce the need of labeled data. We propose the use of autoencoder pre-training in a speaker verification task. First, we t...
Citation
Khan, U.; Hernando, J. DNN speaker embeddings using autoencoder pre-training. A: European Signal Processing Conference. "27th EUSIPCO 2019 European Signal Processing Conference: A Coruña, Spain: September 2-6, 2019". Institute of Electrical and Electronics Engineers (IEEE), 2019, p. 1-5.
Keywords
Autoencoders, Deep learning, Speaker verification, i-vectors
Group of research
IDEAI-UPC - Intelligent Data Science and Artificial Intelligence Research Center
TALP - Centre for Language and Speech Technologies and Applications
VEU - Speech Processing Group

Participants