Gaussian-Bernoulli restricted Boltzmann machines and automatic feature extraction for noise robust missing data mask estimation

A missing data mask estimation method based on Gaussian-Bernoulli restricted Boltzmann machine (GRBM) trained on cross-correlation representation of the audio signal is presented in the study. The automatically learned features by the GRBM are utilized in dividing the time-frequency units of the spe...

Full description

Saved in:
Bibliographic Details
Main Authors: Keronen, Sami, KyungHyun Cho, Raiko, Tapani, Ilin, Alexander, Palomaki, Kalle
Format: Conference Proceeding
Language:English
Subjects:
deep learning
Estimation
Feature extraction
GRBM
mask estimation
Noise
Noise robust
Speech
Speech recognition
Support vector machines
Vectors
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A missing data mask estimation method based on Gaussian-Bernoulli restricted Boltzmann machine (GRBM) trained on cross-correlation representation of the audio signal is presented in the study. The automatically learned features by the GRBM are utilized in dividing the time-frequency units of the spectrographic mask into noise and speech dominant. The system is evaluated against two baseline mask estimation methods in a reverberant multisource environment speech recognition task. The proposed system is shown to provide a performance improvement in the speech recognition accuracy over the previous multifeature approaches.
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2013.6638964