SINGLE CHANNEL JOINT SPEECH DEREVERBERATION AND DENOISING USING DEEP PRIORS

Single channel speech de-reverberation and de-noising is a challenging problem, since directional information is not available in a single channel when compared to multi-channel approaches. Several deep neural network (DNN) based solutions have been proposed in the recent past to solve this problem. These solutions are sequential and de-reverberate the signal after denoising. Additionally these solutions have not utilized the maximum a posteriori (MAP) method which requires the knowledge of the prior. In this work a MAP method is proposed to solve the speech de-reverberation and de-noising problem jointly. A half quadratic splitting (HQS) method is used to solve the joint MAP problem in a DNN framework by splitting it into two minimization problems. The deep prior is modeled using a latent variable and obtained using an iterative method. The performance of the proposed method is illustrated using subjective and objective measures. Experiments on continuous speech recognition are also used to demonstrate the significance of this method.

[1]  Robert M. Gray,et al.  Toeplitz and Circulant Matrices: A Review , 2005, Found. Trends Commun. Inf. Theory.

[2]  Wangmeng Zuo,et al.  Learning Deep CNN Denoiser Prior for Image Restoration , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Simon Doclo,et al.  Speech Dereverberation Using Non-Negative Convolutive Transfer Function and Spectro-Temporal Modeling , 2016, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[4]  Simon King,et al.  The voice bank corpus: Design, collection and data analysis of a large regional accent speech database , 2013, 2013 International Conference Oriental COCOSDA held jointly with 2013 Conference on Asian Spoken Language Research and Evaluation (O-COCOSDA/CASLRE).

[5]  DeLiang Wang,et al.  Robust Speaker Identification in Noisy and Reverberant Conditions , 2014, IEEE/ACM Trans. Audio, Speech & Language Processing.

[6]  Donald Geman,et al.  Nonlinear image recovery with half-quadratic regularization , 1995, IEEE Trans. Image Process..

[7]  Stephen P. Boyd,et al.  Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers , 2011, Found. Trends Mach. Learn..

[8]  Nobutaka Ito,et al.  The Diverse Environments Multi-channel Acoustic Noise Database (DEMAND): A database of multichannel environmental noise recordings , 2013 .

[9]  James R. Glass,et al.  Speech feature denoising and dereverberation via deep autoencoders for noisy reverberant speech recognition , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[10]  Sanjeev Khudanpur,et al.  Librispeech: An ASR corpus based on public domain audio books , 2015, 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[11]  Daniel Povey,et al.  The Kaldi Speech Recognition Toolkit , 2011 .

[12]  DeLiang Wang,et al.  Speech dereverberation and denoising using complex ratio masks , 2017, 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[13]  Xin Wang,et al.  Investigating RNN-based speech enhancement methods for noise-robust Text-to-Speech , 2016 .

[14]  Masato Miyoshi,et al.  Inverse filtering of room acoustics , 1988, IEEE Trans. Acoust. Speech Signal Process..

[15]  Brian Kingsbury,et al.  Recognizing reverberant speech with RASTA-PLP , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[16]  Junichi Yamagishi,et al.  Investigating RNN-based speech enhancement methods for noise-robust Text-to-Speech , 2016, SSW.

[17]  G. K.,et al.  Learning Spectral Mapping for Speech Dereverberation and Denoising , 2017 .

[18]  Antonio Bonafonte,et al.  SEGAN: Speech Enhancement Generative Adversarial Network , 2017, INTERSPEECH.