OlaWod/FreeVC ? reverse-engineered prompt
Reverse engineered prompt
Build me a Python app for one shot voice conversion that can take a short source voice clip and turn it into the voice of a target speaker without needing any text transcript. I want a simple way to run inference from a file list and save the converted audio into an output folder, plus a training flow for the model using prepared audio data. It should support the FreeVC and FreeVC s variants, use the provided pretrained checkpoints if they exist, and be able to work with the dataset prep steps in the repo.
Please make it easy to follow from setup through preprocessing, training, and conversion, and include sensible defaults so I can try it quickly. If anything depends on current model files or setup details, look up the latest docs online if needed.
Are you gonna build this?
make sure you review the code using coderabbit