Wed-1-7-1 NEC-TT Speaker Verification System for SRE'19 CTS Challenge

Kong Aik Lee(Biometrics Research Laboratories, NEC Corporation), Koji Okabe(NEC Corporation), Hitoshi Yamamoto(NEC Corporation), Qiongqiong Wang(Data Science Research Laboratories, NEC Corporation), Ling Guo(Biometrics Research Laboratories, NEC Corporation), Takafumi Koshinaka(Biometrics Research Labs., NEC Corporation), Jiacen Zhang(Tokyo Institute of Technology), Keisuke Ishikawa(Tokyo Institute of Technology) and Koichi Shinoda(Tokyo Institute of Technology)
Abstract: The series of speaker recognition evaluations (SREs) organized by the National Institute of Standards and Technology (NIST) is widely accepted as the de facto benchmark for speaker recognition technology. This paper describes the NEC-TT speaker verification system developed for the recent SRE'19 CTS Challenge. Our system is based on an x-vector embedding front-end followed by a thin scoring back-end. We trained a very-deep neural network for x-vector extraction by incorporating residual connections, squeeze-and-excitation networks, and angular-margin softmax at the output layer. We enhanced the back-end with a tandem approach leveraging the benefit of supervised and unsupervised domain adaptation. We obtained over 30% relative reduction in error rate with each of these enhancements at the front-end and back-end, respectively.
Student Information

Student Events

Travel Grants