Neural Speaker Embeddings with Uncertainty
2023, Disentangling Voice and Content with Self-Supervision [NeurIPS]