Stephen Shum

speaker verification (2) federated learning (1) differential privacy (1) metric learning (1) multi-task learning (1) self-supervised learning (1) knowledge distillation (1) privacy-preserving learning (1) speaker recognition (1) data efficiency (1) parameter efficiency (1) similarity score (1) downstream model simplification (1) voice trigger detection (1) model compression (1) personalized embedding (1)