Expectation-Maximization for Speech Source Separation using Convolutive Transfer Function

Xiaofei Li 1 Laurent Girin 2, 1 Radu Horaud 1
1 PERCEPTION - Interpretation and Modelling of Images and Videos
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
2 GIPSA-CRISSP - CRISSP
GIPSA-DPC - Département Parole et Cognition
Abstract : 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
Complete list of metadatas

Cited literature [30 references]  Display  Hide  Download

https://hal.inria.fr/hal-01982250
Contributor : Team Perception <>
Submitted on : Tuesday, January 15, 2019 - 3:10:49 PM
Last modification on : Monday, April 29, 2019 - 11:52:34 AM
Long-term archiving on : Tuesday, April 16, 2019 - 3:13:27 PM

File

ctf_em.pdf
Files produced by the author(s)

Identifiers

Citation

Xiaofei Li, Laurent Girin, Radu Horaud. Expectation-Maximization for Speech Source Separation using Convolutive Transfer Function. CAAI Transactions on Intelligent Technologies, Elsevier B.V., 2019, 4 (1), pp.47 - 53. ⟨10.1049/trit.2018.1061⟩. ⟨hal-01982250⟩

Share

Metrics

Record views

147

Files downloads

212