DocumentCode
1917765
Title
A frequency domain method for speech separation in a reverberant room
Author
Mischie, Septimiu ; Simion, Georgiana
Author_Institution
Fac. of Electron. & Telecommun., Politeh. Univ. of Timisoara, Timisoara, Romania
fYear
2010
fDate
11-12 Nov. 2010
Firstpage
303
Lastpage
306
Abstract
The paper presents an experimental implementation of a speech separation system. Two sources (loudspeakers) and two sensors in a reverberant room are considered. First, the room impulse responses from each source to each sensor are determined. Then, two speech signals are played to the loudspeakers and the two mixtures are recorded by the two sensors. Using the Fourier transforms of the mixtures and room impulse responses, the original speech signals are separated.
Keywords
Fourier transforms; frequency-domain analysis; loudspeakers; reverberation; source separation; speech processing; transient response; Fourier transforms; frequency domain method; impulse responses; loudspeakers; reverberant room; speech signal separation; Fourier transforms; Loudspeakers; Microphones; Sensors; Signal to noise ratio; Speech; Speech processing; reverberation; room impulse response; speech separation;
fLanguage
English
Publisher
ieee
Conference_Titel
Electronics and Telecommunications (ISETC), 2010 9th International Symposium on
Conference_Location
Timisoara
Print_ISBN
978-1-4244-8457-7
Type
conf
DOI
10.1109/ISETC.2010.5679271
Filename
5679271
Link To Document