Noise-robust linear prediction cepstral features for network speech recognition

View/ Open
File version
Version of Record (VoR)
Author(s)
Alatwi, Aadel
So, Stephen
Paliwal, Kuldip
Year published
2016
Metadata
Show full item recordAbstract
In this paper, we propose a perceptually-motivated method for
modifying the speech power spectrum to obtain a set of linear
prediction coding (LPC) parameters that possess good noiserobustness
properties in network speech recognition. Speech
recognition experiments were performed to compare the accuracy
obtained from MFCC features extracted from AMR-coded
speech that use these modified LPC parameters, as well as from
LPCCs extracted from AMR bitstream parameters. The results
show that when using the proposed LP analysis method, the
recognition performance was on average 1.2% - 6.1% better
than when using the conventional LP ...
View more >In this paper, we propose a perceptually-motivated method for modifying the speech power spectrum to obtain a set of linear prediction coding (LPC) parameters that possess good noiserobustness properties in network speech recognition. Speech recognition experiments were performed to compare the accuracy obtained from MFCC features extracted from AMR-coded speech that use these modified LPC parameters, as well as from LPCCs extracted from AMR bitstream parameters. The results show that when using the proposed LP analysis method, the recognition performance was on average 1.2% - 6.1% better than when using the conventional LP method, depending on the recognition task.
View less >
View more >In this paper, we propose a perceptually-motivated method for modifying the speech power spectrum to obtain a set of linear prediction coding (LPC) parameters that possess good noiserobustness properties in network speech recognition. Speech recognition experiments were performed to compare the accuracy obtained from MFCC features extracted from AMR-coded speech that use these modified LPC parameters, as well as from LPCCs extracted from AMR bitstream parameters. The results show that when using the proposed LP analysis method, the recognition performance was on average 1.2% - 6.1% better than when using the conventional LP method, depending on the recognition task.
View less >
Conference Title
Proceedings of the Sixteenth Australasian International Conference on Speech Science and Technology
Publisher URI
Copyright Statement
© 2016 ASSTA. The attached file is reproduced here in accordance with the copyright policy of the publisher. Please refer to the conference's website for access to the definitive, published version.
Subject
Signal Processing