Title :
Noisy Speech Recognition Based on Robust End-point Detection and Model Adaptation
Author :
Zhang, Zhipeng ; Furui, Sadaoki
Author_Institution :
Multimedia Labs., NTT DoCoMo, Kanagawa, Japan
fDate :
March 18-23, 2005
Keywords :
adaptive signal processing; hidden Markov models; piecewise linear techniques; speech recognition; trees (mathematics); HMM adaptation; best matching GMM selection; blind segmentation window shifting; blind speech segmentation; end-point detection; language model; noise characteristics temporal variation; noisy speech recognition; online model adaptation; piecewise linear-transformation method; robust end-point detection; speech period detection; tree-structured HMM; Acoustical engineering; Adaptation model; Hidden Markov models; Natural languages; Noise level; Noise robustness; Speech enhancement; Speech processing; Speech recognition; Working environment noise;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on
Print_ISBN :
0-7803-8874-7
DOI :
10.1109/ICASSP.2005.1415145
