Paper Info
Title
Adaptive noise immunity learning for word spotting
Abstract
We have developed a word spotting method using adaptive noise immunity learning to achieve robust performance in noisy environments. Previously we proposed a noise immunity word spotting method based on the multiple similarity (MS) method. We also proposed a keyword-based spontaneous speech understanding method and constructed a real-time speech dialogue system, TOSBURG II, based on the noise immunity keyword spotting. By enhancing noise robustness against high-level nonstationary noise in various locations, we have extended noise immunity learning so that it can adapt to noise on-line. In the adaptation process, noisy speech data is synthesized by contaminating pure speech data in a speech database with on-line pick-up background noise data. The synthesized data is then used to modify word reference vectors to adapt to a time-variant noise environment. Massively parallel computers enable real-time adaptation for noise immunity learning. Experimental results indicate the effectiveness of the proposed method
Year
DOI
Venue
1994
10.1109/ICASSP.1994.389259
Acoustics, Speech, and Signal Processing, 1994. ICASSP-94., 1994 IEEE International Conference
Keywords
Field
DocType
adaptive systems,learning systems,noise,parallel processing,speech recognition,speech recognition equipment,TOSBURG II,adaptive noise immunity learning,keyword spotting,massively parallel computers,multiple similarity method,noise robustness,noisy environments,noisy speech data,nonstationary noise,on-line pick-up,real-time speech dialogue system,robust performance,speech data,speech database,spontaneous speech understanding method,synthesized data,time-variant noise environment,word reference vectors
Background noise,Noise measurement,Pattern recognition,Adaptive system,Computer science,Massively parallel,Speech recognition,Keyword spotting,Robustness (computer science),Artificial intelligence,Noise immunity,Spotting
Conference
Volume
ISSN
ISBN
i
1520-6149
0-7803-1775-0
Citations 
PageRank 
References 
0
0.34
9
Authors
2
Name
Order
Citations
PageRank
Yoichi Takebayashi14413.40
Hiroshi Kanazawa2194.70