Abstract | ||
---|---|---|
This paper describes our recent efforts in exploring effective discriminative features for speaker recognition. Recent researches have indicated that the appropriate fusion of features is critical to improve the performance of speaker recognition system. In this paper we describe our approaches for the NIST 2006 Speaker Recognition Evaluation. Our system integrated the cepstral GMM modeling, cepstral SVM modeling and tokenization at both phone level and frame level. The experimental results on both NIST 2005 SRE corpus and NIST 2006 SRE corpus are presented. The fused system achieved 8.14% equal error rate on 1conv4w-1conv4w test condition of the NIST 2006 SRE. |
Year | DOI | Venue |
---|---|---|
2006 | 10.1007/11939993_59 | ISCSLP |
Keywords | Field | DocType |
fused system,recent effort,speaker recognition evaluation,speaker recognition,sre corpus,frame level,tokenization feature,cepstral svm modeling,phone level,cepstral gmm modeling,speaker recognition system,system integration,fusion,support vector machine,tokenization,gaussian mixture model | Speech processing,Tokenization (data security),Computer science,Word error rate,Cepstrum,Support vector machine,Speech recognition,NIST,Speaker recognition,Natural language processing,Artificial intelligence,Discriminative model | Conference |
Volume | ISSN | ISBN |
4274 | 0302-9743 | 3-540-49665-3 |
Citations | PageRank | References |
7 | 0.65 | 10 |
Authors | ||
10 |
Name | Order | Citations | PageRank |
---|---|---|---|
Rong Tong | 1 | 108 | 11.33 |
Bin Ma | 2 | 44 | 4.45 |
Kong-Aik Lee | 3 | 709 | 60.64 |
Changhuai You | 4 | 40 | 4.10 |
Donglai Zhu | 5 | 117 | 13.59 |
Tomi Kinnunen | 6 | 1323 | 86.67 |
Hanwu Sun | 7 | 98 | 14.15 |
Minghui Dong | 8 | 201 | 33.61 |
Eng Siong Chng | 9 | 970 | 106.33 |
Haizhou Li | 10 | 3678 | 334.61 |