Ahmadzei's picture
added 3 more tables for large emb model
5fa1a76
[ClvpConditioningEncoder] takes those text tokens and audio representations and converts them into embeddings conditioned on the text and audio.