Title :
Speech segregation based on pitch tracking and amplitude modulation
Author :
Hu, Guoning ; Wang, DeLiang
Author_Institution :
Biophys. Program, Ohio State Univ., Columbus, OH, USA
Abstract :
Speech segregation is an important task of auditory scene analysis (ASA), in which the speech of a certain speaker is separated from other interfering signals. D.L. Wang and G.J. Brown (see IEEE Trans. Neural Network, vol.10, p.684-97, 1999) proposed a multistage neural model for speech segregation, the core of which is a two-layer oscillator network. We extend their model by adding further processes based on psychoacoustic evidence to improve the performance. These processes include pitch tracking and grouping based on amplitude modulation (AM). Our model is systematically evaluated and compared with the Wang-Brown model, and it yields significantly better performance
Keywords :
amplitude modulation; neural nets; parameter estimation; speech processing; tracking; amplitude modulation; auditory scene analysis; auditory system; multistage neural model; pitch contour estimation; pitch grouping; pitch tracking; psychoacoustic evidence; speech segregation; two-layer neural oscillator network; Amplitude modulation; Automatic speech recognition; Biophysics; Image analysis; Modulation coding; Oscillators; Psychoacoustic models; Psychology; Speech analysis; Speech processing;
Conference_Titel :
Applications of Signal Processing to Audio and Acoustics, 2001 IEEE Workshop on the
Conference_Location :
New Platz, NY
Print_ISBN :
0-7803-7126-7
DOI :
10.1109/ASPAA.2001.969547
