LIA system description for NIST SRE 2016
This is an incremental improvement for speaker recognition in evaluation campaigns.
The paper describes the LIA speaker recognition system for the NIST SRE 2016 evaluation, which uses eight sub-systems based on i-vector/PLDA techniques with variations in feature extraction, i-vector extraction, and data-shifting, and fuses them at the score-level.
This paper describes the LIA speaker recognition system developed for the Speaker Recognition Evaluation (SRE) campaign. Eight sub-systems are developed, all based on a state-of-the-art approach: i-vector/PLDA which represents the mainstream technique in text-independent speaker recognition. These sub-systems differ: on the acoustic feature extraction front-end (MFCC, PLP), at the i-vector extraction stage (UBM, DNN or two-feats posteriors) and finally on the data-shifting (IDVC, mean-shifting). The submitted system is a fusion at the score-level of these eight sub-systems.