Conference paper

 
Motlíček, P., Černocký, J.: Time-domain based Temporal Processing with Application of, In: Proc. EUROSPEECH 2003, Geneva, CH, IDIAP, 2003, p. 821-824, ISSN 1018-4074
Publication language:english
Original title:Time-domain based Temporal Processing with Application of
Title (cs):Zpracování řečového signálu v časové oblsati s použitím ortogonálních transformací
Pages:821-824
Proceedings:Proc. EUROSPEECH 2003
Conference:Eurospeech 2003-Switzerland - 8th European conference on speech communication and technology
Place:Geneva, CH
Year:2003
Journal:European Speech Communication, Vol. 2003, No. 9, CZ
ISSN:1018-4074
Publisher:Institute for Perceptual Artificial Intelligence
URL:http://www.symporg.ch/eurospeech/ [HTML]
URL:http://www.fit.vutbr.cz/~motlicek/publi/2003/motlicek_02.pdf [PDF]
Keywords
speech proceesing, speech recognition, TRAP, feature extraction
Annotation
Time-domain based Temporal Processing with Application of Orthogonal Transformations
Abstract
In the paper, novel approach that efficiently extracts the temporal information of speech has been proposed. This algorithm is fully employed in time-domain, and the preprocessing blocks are well justified by psychoacoustic studies. The achieved results show the different properties of proposed algorithm compared to the traditional approach. The algorithm is advantageous in terms of possible modifications and computational inexpensiveness. Then, in our experiments, we have focused on different representation of time trajectories. Classical methods that are efficient in conventional feature extraction approaches showed not to be suitable to approximate temporal trajectories of speech. However, the application of some orthogonal transformations, such as discrete Fourier transform or discrete cosine transform, on top of previously derived temporal trajectories outperforms classification in original domain. In addition, these transformed features are very efficient to reduce the dimensionality of data. %in data reduction.
BibTeX:
@INPROCEEDINGS{
   author = {Petr Motlíček and Jan Černocký},
   title = {Time-domain based Temporal Processing with Application of},
   pages = {821--824},
   booktitle = {Proc. EUROSPEECH 2003},
   journal = {European Speech Communication},
   volume = {2003},
   number = {9},
   year = {2003},
   location = {Geneva, CH},
   publisher = {Institute for Perceptual Artificial Intelligence},
   ISSN = {1018-4074},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php?id=7232}
}