DocumentCode
1874253
Title
Robust digit recognition using phase-dependent time-frequency masking
Author
Shi, Guangji ; Aarabi, Parham
Author_Institution
Dept. of Electr. & Comput. Eng., Toronto Univ., Ont., Canada
Volume
3
fYear
2003
fDate
6-9 July 2003
Abstract
A technique using the time-frequency phase information of two microphones is proposed to estimate an ideal time-frequency mask using time-delay-of-arrival (TDOA) of the signal of interest. At a signal-to-noise ratio (SNR) of 0dB, the proposed technique using two microphones achieves a digit recognition rate (average over 5 speakers, each speaking 20-30 digits) of 71%. In contrast, delay-and-sum beamforming only achieves a 40% recognition rate with two microphones and 60% with four microphones. Superdirective beamforming achieves a 44% recognition rate with two microphones and 65% with four microphones.
Keywords
microphones; speech intelligibility; speech recognition; time-frequency analysis; delay-and-sum beamforming; ideal time-frequency mask; microphones; phase-dependent time-frequency masking; robust digit recognition; signal-to-noise ratio; speech recognition; superdirective beamforming; time-delay-of-arrival; time-frequency phase information; Array signal processing; Delay; Frequency domain analysis; Gaussian noise; Independent component analysis; Microphones; Robustness; Speech enhancement; Speech recognition; Time frequency analysis;
fLanguage
English
Publisher
ieee
Conference_Titel
Multimedia and Expo, 2003. ICME '03. Proceedings. 2003 International Conference on
Print_ISBN
0-7803-7965-9
Type
conf
DOI
10.1109/ICME.2003.1221390
Filename
1221390
Link To Document