JSTSP Volume 13 Issue 4

This paper presents a time-frequency masking based online multi-channel speech enhancement approach that uses a convolutional recurrent neural network to estimate the mask. The magnitude and phase components of the short-time Fourier transform coefficients for multiple time frames are provided as an input such that the network is able to discriminate between the directional speech...

The seven papers in this special issue cover various far-field speech processing techniques including speech enhancement, separation and recognition, and their integration. In most of the methods, multichannel speech processing is an essential component to achieve state-of-the-art performance.