Paper Info
Title
Medium-duration modulation cepstral feature for robust speech recognition
Abstract
Studies have shown that the performance of state-of-the-art automatic speech recognition (ASR) systems significantly deteriorate with increased noise levels and channel degradations, when compared to human speech recognition capability. Traditionally, noise-robust acoustic features are deployed to improve speech recognition performance under varying background conditions to compensate for the performance degradations. In this paper, we present the Modulation of Medium Duration Speech Amplitude (MMeDuSA) feature, which is a composite feature capturing subband speech modulations and a summary modulation. We analyze MMeDuSA's speech recognition performance using SRI International's DECIPHER® large vocabulary continuous speech recognition (LVCSR) system, on noise and channel degraded Levantine Arabic speech distributed through the Defense Advance Research Projects Agency (DARPA) Robust Automatic Speech Transcription (RATS) program. We also analyzed MMeDuSA's performance against the Aurora-4 noise-and-channel degraded English corpus. Our results from all these experiments suggest that the proposed MMeDuSA feature improved recognition performance under both noisy and channel degraded conditions in almost all the recognition tasks.
Year
DOI
Venue
2014
10.1109/ICASSP.2014.6853898
Acoustics, Speech and Signal Processing
Keywords
Field
DocType
amplitude modulation,cepstral analysis,speech processing,speech recognition,ASR system,DARPA,Decipher,LVCSR system,MMeDuSA feature,RATS program,SRI international,aurora-4 noise-and-channel degraded English corpus,automatic speech recognition system,cepstral feature,channel degradation,channel degraded Levantine Arabic speech,composite feature capturing subband speech modulation,defense advance research projects agency,human speech recognition capability,large vocabulary continuous speech recognition,medium-duration modulation,modulation of medium duration speech amplitude,noise level,noise-robust acoustic feature,robust automatic speech transcription,robust speech recognition,speech recognition performance,summary modulation,large vocabulary continuous speech recognition,modulation features,noise-robust speech recognition
Speech processing,Speech coding,Pattern recognition,Computer science,Voice activity detection,Cepstrum,Communication channel,Speech recognition,Speaker recognition,Artificial intelligence,Linear predictive coding,Acoustic model
Conference
ISSN
Citations 
PageRank 
1520-6149
17
0.76
References 
Authors
9
4
Name
Order
Citations
PageRank
Vikramjit Mitra129924.83
Horacio Franco254372.04
Martin Graciarena328124.70
Dimitra Vergyri437336.97