DocumentCode :
180089
Title :
Unsupervised domain adaptation for deep neural network based voice activity detection
Author :
Xiao-Lei Zhang
Author_Institution :
Dept. of Electron. Eng., Tsinghua Univ., Beijing, China
fYear :
2014
fDate :
4-9 May 2014
Firstpage :
6864
Lastpage :
6868
Abstract :
The mismatching problem between the training and test speech corpora hinders the practical use of the machine-learning-based voice activity detection (VAD). In this paper, we try to address this problem by the unsupervised domain adaptation techniques, which try to find a shared feature subspace between the mismatching corpora. The denoising deep neural network is used as the learning machine. Three domain adaptation techniques are used for analysis. Experimental results show that the unsupervised domain adaptation technique is promising to the mismatching problem of VAD.
Keywords :
neural nets; speech recognition; unsupervised learning; deep neural network; learning machine; unsupervised domain adaptation; voice activity detection; Acoustics; Adaptation models; Noise; Speech; Speech processing; Speech recognition; Training; deep learning; domain adaptation; feature learning; transfer learning; voice activity detection;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2014 IEEE International Conference on
Conference_Location :
Florence
Type :
conf
DOI :
10.1109/ICASSP.2014.6854930
Filename :
6854930
Link To Document :
بازگشت