Single-channel speech separation using zero-phase models

Abstract

This paper addresses the problem of single-channel speech separation to extract and enhance the desired speech signals from mixed speech signals. We propose a new speech separation algorithm by utilizing both magnitude and phase information, which can be applied to multimedia mobile communication and navigation systems. Conventionally, phase information has been neglected in speech signal processing. However, in the proposed method, we originally formulate a probabilistic phase-based speech estimator based on zero-phase models to improve the speech separation performance. In the speech separation experiments, the proposed method is shown to improve speaker-to-interference ratio (SIR) by 2.2 dB compared to the system using magnitude models only. When only phase-based speech estimator is used for speech separation, the SIR was improved by 0.8 dB. This result justify that the proposed phase-based speech estimation method achieves significant SIR improvement compared with the previous magnitude-based method.