joonson / syncnet_trainer

Disentangled Speech Embeddings using Cross-Modal Self-Supervision
MIT License
155 stars 26 forks source link