DocumentCode :
1654183
Title :
Denoising deep neural networks based voice activity detection
Author :
Xiao-Lei Zhang ; Ji Wu
Author_Institution :
Dept. of Electron. Eng., Tsinghua Univ., Beijing, China
fYear :
2013
Firstpage :
853
Lastpage :
857
Abstract :
Recently, the deep-belief-networks (DBN) based voice activity detection (VAD) has been proposed. It is powerful in fusing the advantages of multiple features, and achieves the state-of-the-art performance. However, the deep layers of the DBN-based VAD do not show an apparent superiority to the shallower layers. In this paper, we propose a denoising-deep-neural-network (DDNN) based VAD to address the aforementioned problem. Specifically, we pre-train a deep neural network in a special unsupervised denoising greedy layer-wise mode, and then fine-tune the whole network in a supervised way by the common back-propagation algorithm. In the pre-training phase, we take the noisy speech signals as the visible layer and try to extract a new feature that minimizes the reconstruction cross-entropy loss between the noisy speech signals and its corresponding clean speech signals. Experimental results show that the proposed DDNN-based VAD not only outperforms the DBN-based VAD but also shows an apparent performance improvement of the deep layers over shallower layers.
Keywords :
backpropagation; belief networks; greedy algorithms; neural nets; signal denoising; speech processing; unsupervised learning; DBN based VAD; DDNN based VAD; DDNN-based VAD; common back-propagation algorithm; deep layers; deep-belief-networks; denoising deep neural networks; denoising-deep-neural-network based VAD; shallower layers; special unsupervised denoising greedy layer-wise mode; state-of-the-art performance; voice activity detection; Feature extraction; Neural networks; Noise; Noise measurement; Noise reduction; Speech; Training; Deep learning; denoising deep neural networks; voice activity detection;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference on
Conference_Location :
Vancouver, BC
ISSN :
1520-6149
Type :
conf
DOI :
10.1109/ICASSP.2013.6637769
Filename :
6637769
Link To Document :
بازگشت