Paper Info
Title
Improved low bit-rate audio compression using reduced rank ICA instead of psychoacoustic modeling
Abstract
Traditional audio coding is based on a perceptual compres- sion paradigm that exploits psychoacoustic information to efficiently encode audio signals. Recently, extensive re- search has been conducted in order to understand how the brain encodes natural signals. These results suggest that the encoding process is very efficient in terms of redundancy reduction of the signal information. It could be that the psy- choacoustic effects (such as the masking effect) are only a special case of a more general redundancy reduction mech- anism that exists in the auditory pathway. Motivated by this work we propose a new audio coding scheme that is based on improved sound representation found by Indepen- dent Component Analysis. Using a local linear, low rank, non-orthogonal transform, we remove additional redundan- cies in the signal. At low bitrates this coding scheme gives results superior to a legacy perceptual encoding scheme for different kinds of audio signals.
Year
DOI
Venue
2003
10.1109/ICASSP.2003.1200006
ICASSP '03). 2003 IEEE International Conference
Keywords
Field
DocType
audio coding,brain,data compression,independent component analysis,redundancy,audio coding,audio signals,auditory pathway,brain,encoding process,improved sound representation,independent component analysis,local linear nonorthogonal transform,low bit-rate audio compression,low rank nonorthogonal transform,natural signals,reduced rank ICA,redundancy reduction
Audio signal,Speech coding,Pattern recognition,Computer science,Transform coding,Sound quality,Speech recognition,Joint (audio engineering),Artificial intelligence,Sub-band coding,Data compression,Dynamic range compression
Conference
Volume
ISSN
ISBN
5
1520-6149
0-7803-7663-3
Citations 
PageRank 
References 
3
0.50
2
Authors
3
Name
Order
Citations
PageRank
Adiel Ben-Shalom130.50
M. Werman2343112.04
Shlomo Dubnov328345.23