Acoustic backing-off as an implementation of missing feature theory

DOI10.1016/S0167-6393(00)00037-6MaRDI QIDQ4794933zbMATH OpenFDO

Authors Johan de Veth, Bert Cranen, Louis Boves

Publication date 20 February 2003

Published in Speech Communication (Search for Journal in Brave)

zbMATH Keywords

automatic speech recognitionacoustic backing-offmismatched training-test conditionsmissing feature theory

Mathematics Subject Classification ID

Pattern recognition, speech recognition (68T10) Computing methodologies and applications (68U99)

Recommendations

• Acoustic features and a distance measure that reduce the impact of training–test mismatch in ASR
• Acoustic features and distance measure to reduce vulnerability of ASR performance due to the presence of communication channel and/or background noise
• Uncertainty decoding and conditional Bayesian estimation
• Robust automatic speech recognition with missing and unreliable acoustic data
• Distortion-class modeling for robust speech recognition under GSM RPE-LTP coding

Cited in

(5)

• Distortion-class modeling for robust speech recognition under GSM RPE-LTP coding
• Acoustic features and distance measure to reduce vulnerability of ASR performance due to the presence of communication channel and/or background noise
• Acoustic features and a distance measure that reduce the impact of training–test mismatch in ASR
• Robust automatic speech recognition with missing and unreliable acoustic data
• Reconstructing noise-corrupted spectrographic components for robust speech recognition

This page was built for publication: Acoustic backing-off as an implementation of missing feature theory

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4794933)