Abstract | ||
---|---|---|
In this paper, we present a pseudo-key analysis approach for cross-validation of language recognition systems before the ground truth (true key) becomes available. A state-of-the-art language recognition system typically employs multiple language recognition classifiers which are fused to form a mixture of experts. The individual classifiers are also called subsystems. To avoid the fused system from being brought down by some outlier classifiers, pseudo keys are designed to cross-examine the integrity of individual classifier candidates. The language recognition experiments are conducted on the NIST 2007 Language Recognition Evaluation (LRE) corpus using the subsystems in the primary submission from the Institute for Infocomm Research (IIR). |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/ICASSP.2009.4960593 | ICASSP |
Keywords | Field | DocType |
pseudo key,infocomm research,language recognition experiment,fused system,ground truth,individual classifier candidate,individual classifier,state-of-the-art language recognition system,language recognition system,language recognition evaluation,multiple language recognition system,multiple language recognition classifier,data mining,feature extraction,sun,support vector machines,cross validation,kernel,speech recognition,natural language processing,language,natural languages,nist,acoustics,design,speech,telephony | Cache language model,Computer science,Support vector machine,Outlier,Speech recognition,Feature extraction,NIST,Natural language processing,Artificial intelligence,Classifier (linguistics),Cross-validation,Language model | Conference |
ISSN | Citations | PageRank |
1520-6149 | 0 | 0.34 |
References | Authors | |
5 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hanwu Sun | 1 | 98 | 14.15 |
Bin Ma | 2 | 600 | 47.26 |
Haizhou Li | 3 | 3678 | 334.61 |