End-to-End Text-Dependent Speaker Verification

   Abstract