DKU's ResNet-based system with angular softmax and WPE achieves 0.3532 minDCF and 4.96% EER on the VOiCES 2019 evaluation set using cosine scoring.
Exploring the Encoding Layer and Loss Function in End-to-End Speaker and Language Recognition System,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
The DKU System for the Speaker Recognition Task of the 2019 VOiCES from a Distance Challenge
DKU's ResNet-based system with angular softmax and WPE achieves 0.3532 minDCF and 4.96% EER on the VOiCES 2019 evaluation set using cosine scoring.