DEL_Segments

Segments and trials files used in our paper, all utterances are between 3-5s after vad.

• "Joint i-vector with End-to-End system for Short Duration Text-independent speaker verification", ICASSP 2018
• "Angular Softmax for Short-Duration Text-independent Speaker Verification", Interspeech 2018