Abstract | ||
---|---|---|
For text-independent speaker identification a promi- nent combination is to use Gaussian Mixture Models (GMM) for classification while relying on Mel-Frequency Cepstral Co- efficients (MFCC) as features. To take temporal information into account the time difference of features of adjacent speech frames are appended to the initial features. In this paper we investigate the applicability of spectro-temporal features obtained from Gabor-Filters and present an algorithm for optimizing the possible parameters. Simulation results on a database show that spectro-temporal features achieve higher recognition rates than purely temporal features for clean speech as well as for disturbed speech. |
Year | DOI | Venue |
---|---|---|
2007 | 10.1109/ISCAS.2007.378660 | New Orleans, LA |
Keywords | Field | DocType |
Gabor filters,Gaussian processes,cepstral analysis,speaker recognition,Gabor features,Gabor-filters,Gaussian mixture models,Mel-frequency cepstral coefficients,spectro-temporal features,speech frames,text-independent speaker identification | Mel-frequency cepstrum,Speaker identification,Pattern recognition,Computer science,Speech recognition,Speaker recognition,Gaussian process,Artificial intelligence,Cepstral analysis,Time difference,Mixture model | Conference |
ISSN | ISBN | Citations |
0271-4302 | 1-4244-0921-7 | 2 |
PageRank | References | Authors |
0.49 | 4 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Volker Mildner | 1 | 2 | 0.49 |
Stefan Goetze | 2 | 132 | 15.15 |
Kammeyer, K.-D. | 3 | 194 | 21.42 |
Alfred Mertins | 4 | 534 | 76.48 |