Voice Recognition White Papers
Suppression Rule for Speech Recognition Friendly Noise Suppressors
Overview Audio signal enhancement often involves the application of a time-varying filter, or suppression rule, to the frequency-domain transform of a corrupted signal. Known approaches use rules derived under Gaussian models and interpret them as spectral estimators in a Bayesian statistical framework. While this mathematical approach provides rules that satisfy certain optimization criteria these rules are not optimal when the enhanced signal is for a speech recognition engine. This paper presents the approach and the results for creation of a speech recognition friendly suppression rule. The described approach increases the average speech recognition rate in Aurora 2 tests from 52.47% to 77.69% while maintaining performance for low noise utterances.
| Publisher | Microsoft | File Format | |
|---|---|---|---|
| Date Published | December 2005 | ||
| Format | White Papers | ||
| Topics | |||
-
Did you find this white paper useful? 4 out of 6 users found this white paper useful
White Paper RSS
Download
Register
