LEARNING EMOTION-INVARIANT SPEAKER REPRESENTATIONS FOR SPEAKER VERIFICATION

This repository contains the training sets, test sets, and test trails of the paper "LEARNING EMOTION-INVARIANT SPEAKER REPRESENTATIONS FOR SPEAKER VERIFICATION".

Audio files

All audio files come from the open-source dataset Dusha. Audio files can be downloaded from here.

Training list

Training list can be found in the train folder.

Test list

Test list can be found in the test folder.

Test trails

All test trails can be found in the trails folder.