Abstract | ||
---|---|---|
We propose a language recognition system based on discriminative vectors, in which parallel phone recognizers serve as the voice tokenization front-end followed by vector space modeling that effectively vectorizes phonotactic features, and the final classification is carried out based on the discriminative vectors. We design an ensemble of discriminative binary classifiers. The output values of these classifiers construct a discriminative vector, also referred to as output codes, to represent the high-dimensional phonotactic features. We achieve equal-error-rate of 1.95%, 3.02% and 4.9% on 1996, 2003 and 2005 NIST LRE databases, respectively, for 30-second trials. |
Year | DOI | Venue |
---|---|---|
2007 | 10.1109/ICASSP.2007.367241 | ICASSP (4) |
Keywords | Field | DocType |
phonotactic features,speech recognition,discriminative vector,ensemble classifiers,discriminative binary classifiers,parallel phone recognizers,output codes,spoken language recognition,statistics,support vector machines,principal component analysis,artificial neural networks,feature extraction,vector space model,front end,nist,natural languages | Tokenization (data security),Pattern recognition,Computer science,Support vector machine,Speech recognition,Feature extraction,NIST,Natural language,Artificial intelligence,Artificial neural network,Discriminative model,Spoken language | Conference |
Volume | ISSN | ISBN |
4 | 1520-6149 | 1-4244-0727-3 |
Citations | PageRank | References |
4 | 0.44 | 7 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Bin Ma | 1 | 600 | 47.26 |
Rong Tong | 2 | 108 | 11.33 |
Haizhou Li | 3 | 3678 | 334.61 |