Getting More for Less: Using Weak Labels and AV-Mixup for Robust Audio-Visual Speaker Verification.

Anith Selvakumar, Homa Fashandi

Browse the full Interspeech paper archive.