[Model Request] Support for YingMusic-SVC (based on Seed-VC)

[mcolletta]

Hi everyone,

First of all, thank you for this amazing project!

I would like to request support for YingMusic-SVC (by GiantAILab), which delivers outstanding results in Singing Voice Conversion.

Architecture & Feasibility

YingMusic-SVC is explicitly built on top of the Seed-VC architecture, which is already integrated into audio.cpp.

Since the core structure is shared, integrating this model might mostly require adapting the pipeline to support its specific training biases and weights, rather than implementing a brand-new backend from scratch.

References

• Official Repository: https://github.com/GiantAILab/YingMusic-SVC
• Hugging Face Hub: https://huggingface.co/GiantAILab/YingMusic-SVC

Thank you for considering this request!