Radboud Universiteit Nijmegen
Single-channel speech separation algorithms frequently ignore the issue of accurate phase estimation while reconstructing the enhanced signal. Instead, they directly employ the mixed-signal phase for signal reconstruction which leads to undesired traces of the interfering source in the target signal. In this paper, assuming a given knowledge of signal spectrum amplitude, the authors present a solution to estimate the phase information for signal reconstruction of the sources from a single-channel mixture observation. They first investigate the effectiveness of the proposed phase estimation method employing known magnitude spectra of sources as an ideal case. They further relax the ideal signal spectra assumption by perturbing the clean signal spectra via Gaussian noise.