Application of shape analysis techniques for improved CASA-based speech separation

This paper addresses the problem of single-channel speech separation for robust speech recognition in noisy environments. We propose to apply shape analysis techniques to a speech separation system based on computational auditory scene analysis (CASA). The system extracts the desired speech signals from a mixture of various sound sources with an additional module to complement missing speech signals. In the speech separation experiment using various artificially-mixed speech and noise sources, the proposed system increases the signal-to-noise ratio by 6.6 dB. In the speech recognition experiment using the Interspeech speech database, it improves recognition accuracy by 22%.