Expectation-Maximization for Speech Source Separation Using Convolutive Transfer Function

10 Apr 2019  ·  Li Xiaofei, Girin Laurent, Horaud Radu ·

This paper addresses the problem of under-determinded speech source separation from multichannel microphone singals, i.e. the convolutive mixtures of multiple sources. The time-domain signals are first transformed to the short-time Fourier transform (STFT) domain... To represent the room filters in the STFT domain, instead of the widely-used narrowband assumption, we propose to use a more accurate model, i.e. the convolutive transfer function (CTF). At each frequency band, the CTF coefficients of the mixing filters and the STFT coefficients of the sources are jointly estimated by maximizing the likelihood of the microphone signals, which is resolved by an Expectation-Maximization (EM) algorithm. Experiments show that the proposed method provides very satisfactory performance under highly reverberant environments. read more

PDF Abstract
No code implementations yet. Submit your code now


Sound Audio and Speech Processing


  Add Datasets introduced or used in this paper