Boosted deep neural networks and multi-resolution cochleagram features for voice activity detection

Voice activity detection (VAD) is an important frontend of many speech processing systems. In this paper, we describe a new VAD algorithm based on boosted deep neural networks (bDNNs). The proposed algorithm first generates multiple base predictions for a single frame from only one DNN and then aggregates the base predictions for a better prediction of the frame. Moreover, we employ a new acoustic feature, multi-resolution cochleagram (MRCG), that concatenates the cochleagram features at multiple spectrotemporal resolutions and shows superior speech separation results over many acoustic features. Experimental results show that bDNN-based VAD with the MRCG feature outperforms state-of-the-art VADs by a considerable margin.

[1]  Marc'Aurelio Ranzato,et al.  Large Scale Distributed Deep Networks , 2012, NIPS.

[2]  Mark Liberman,et al.  Speech activity detection on youtube using deep neural networks , 2013, INTERSPEECH.

[3]  Wonyong Sung,et al.  A statistical model-based voice activity detection , 1999, IEEE Signal Processing Letters.

[4]  Yunde Jia,et al.  Voice Activity Detection Via Noise Reducing Using Non-Negative Sparse Coding , 2013, IEEE Signal Processing Letters.

[5]  DeLiang Wang,et al.  A feature study for classification-based speech separation at very low signal-to-noise ratio , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[6]  Guy J. Brown,et al.  Computational Auditory Scene Analysis: Principles, Algorithms, and Applications , 2006 .

[7]  Geoffrey E. Hinton,et al.  On the importance of initialization and momentum in deep learning , 2013, ICML.

[8]  Xiao-Lei Zhang,et al.  Deep Belief Networks Based Voice Activity Detection , 2013, IEEE Transactions on Audio, Speech, and Language Processing.

[9]  Joon-Hyuk Chang,et al.  Voice activity detection based on statistical models and machine learning approaches , 2010, Comput. Speech Lang..

[10]  Sridhar Krishna Nemala,et al.  A Multistream Feature Framework Based on Bandpass Modulation Filtering for Robust Speech Recognition , 2013, IEEE Transactions on Audio, Speech, and Language Processing.

[11]  John H. L. Hansen,et al.  Discriminative Training for Multiple Observation Likelihood Ratio Based Voice Activity Detection , 2010, IEEE Signal Processing Letters.

[12]  Hoirin Kim,et al.  Multiple Acoustic Model-Based Discriminative Likelihood Ratio Weighting for Voice Activity Detection , 2012, IEEE Signal Processing Letters.

[13]  DeLiang Wang,et al.  Auditory Segmentation Based on Onset and Offset Analysis , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[14]  Jianwu Dang,et al.  Voice Activity Detection Based on an Unsupervised Learning Framework , 2011, IEEE Transactions on Audio, Speech, and Language Processing.

[15]  John H. L. Hansen,et al.  Robust front-end processing for speaker identification over extremely degraded communication channels , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[16]  Thad Hughes,et al.  Recurrent neural networks for voice activity detection , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[17]  Athanasios Katsamanis,et al.  Multi-band long-term signal variability features for robust voice activity detection , 2013, INTERSPEECH.

[18]  Israel Cohen,et al.  Voice Activity Detection in Presence of Transient Noise Using Spectral Clustering , 2013, IEEE Transactions on Audio, Speech, and Language Processing.

[19]  Javier Ramírez,et al.  Statistical voice activity detection using a multiple observation likelihood ratio test , 2005, IEEE Signal Processing Letters.

[20]  Ramjee Prasad,et al.  Convex Combination of Multiple Statistical Models With Application to VAD , 2011, IEEE Transactions on Audio, Speech, and Language Processing.

[21]  Yoshihiko Nankaku,et al.  Voice activity detection based on conditional random fields using multiple features , 2010, INTERSPEECH.

[22]  Dong Yu,et al.  Deep-structured hidden conditional random fields for phonetic recognition , 2010, INTERSPEECH.

[23]  Thomas G. Dietterich.  Multiple Classifier Systems , 2000, Lecture Notes in Computer Science.

[24]  Tara N. Sainath,et al.  Improving deep neural networks for LVCSR using rectified linear units and dropout , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.