AIbase
Product LibraryTool Navigation

dis-Vector-Embedding

Public

The Dis-Vector project enhances voice conversion and synthesis through disentangled embeddings, allowing for high-quality, zero-shot voice cloning across multiple languages. This model leverages separate encoders for content, pitch, rhythm, and timbre, enabling precise control over synthesized voice characteristics.

Creat2024-04-10T14:16:13
Update2025-03-15T23:17:28
https://nn-project-1.github.io/dis-vector_web/
1
Stars
0
Stars Increase