Abstract | ||
---|---|---|
A novel method for speech recognition is presented, utiliz- ing nonlinear/chaotic signal processing techniques to extract time-domain based, reconstructed phase space features. This work examines the incorporation of trajectory information into this model as well as the combination of both MFCC and RPS feature sets into one joint feature vector. The results demon- strate that integration of trajectory information increases the recognition accuracy of the typical RPS feature set, and when MFCC and RPS feature sets are combined, improvement is made over the baseline. This result suggests that the features extracted using these nonlinear techniques contain different discriminatory information than the features extracted from linear approaches alone. |
Year | DOI | Venue |
---|---|---|
2004 | 10.1109/ICASSP.2004.1326040 | ICASSP '04). IEEE International Conference |
Keywords | Field | DocType |
chaos,feature extraction,signal reconstruction,speech recognition,time-frequency analysis,MFCC,RPS feature sets,discriminatory information,feature extraction,frequency domain features,joint feature vector,nonlinear/chaotic signal processing,recognition accuracy,reconstructed phase space features,speech recognition,time-domain based features,trajectory information | Frequency domain,Signal processing,Mel-frequency cepstrum,Feature vector,Pattern recognition,Computer science,Feature (computer vision),Feature extraction,Speech recognition,Feature (machine learning),Artificial intelligence,Signal reconstruction | Conference |
Volume | ISSN | ISBN |
1 | 1520-6149 | 0-7803-8484-9 |
Citations | PageRank | References |
5 | 0.44 | 1 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Andrew C. Lindgren | 1 | 79 | 4.99 |
Michael T. Johnson | 2 | 435 | 53.51 |
Richard J. Povinelli | 3 | 13 | 1.06 |