DocumentCode :
431285
Title :
Language Model Estimation for Optimizing End-to-end Performance of a Natural Language Call Routing System
Author :
Goel, Vaibhava ; Kuo, Hong-Kwang Jeff ; Deligne, Sabine ; Wu, Cheng
Author_Institution :
IBM T. J. Watson Res. Center, Yorktown Heights, NY, USA
Volume :
1
fYear :
2005
fDate :
March 18-23, 2005
Firstpage :
565
Lastpage :
568
Keywords :
error statistics; interactive systems; maximum likelihood estimation; natural language interfaces; optimisation; speech recognition; speech-based user interfaces; unsupervised learning; LM updates; WER; automatic speech recognition; classification accuracy; conditional maximum likelihood estimation; end-to-end performance; extended Baum-Welch procedure; language model estimation; natural language call routing system; optimization; spoken dialogue systems; statistical models; training; unsupervised learning; untranscribed speech data; word error rate; Automatic speech recognition; Error analysis; Maximum likelihood estimation; Natural languages; Routing; Speech recognition; System performance; Telephony; Training data; Unsupervised learning;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on
ISSN :
1520-6149
Print_ISBN :
0-7803-8874-7
Type :
conf
DOI :
10.1109/ICASSP.2005.1415176
Filename :
1415176
Link To Document :
بازگشت