Paper Info
Title
Double the trouble: handling noise and reverberation in far-field automatic speech recognition
Abstract
Far-field microphone speech signals cause high error rates for automatic speech recognition systems, due to room reverberation and lower signal-to-noise ratios. We have ob- served large increases in speech recognition word error rates when using a far-field (3-6 feet) microphone in a conference room, in comparison with recordings from close-talking mi- crophones. In an earlier paper, we showed improvements in far-field speech recognition performance using a long- term log spectral subtraction method to combat reverbera- tion. This method is based on a principle similar to cepstral mean subtraction but uses a much longer analysis window (e.g., 1 s) in order to deal with reverberation. Here we show that a combination of short-term noise filtering and long- term log spectral subtraction can further reduce recognition word error rates.
Year
Venue
Keywords
2002
INTERSPEECH
speech recognition,error rate,automatic speech recognition,signal to noise ratio,word error rate
Field
DocType
Citations 
Speech processing,Reverberation,Pattern recognition,Voice activity detection,Computer science,Conference room,Near and far field,Filter (signal processing),Speech recognition,Artificial intelligence,Speech recognition performance,Microphone
Conference
13
PageRank 
References 
Authors
1.77
6
2
Name
Order
Citations
PageRank
David Gelbart113417.54
Nelson Morgan23048533.52