OpenVoice
OpenVoice copied to clipboard
The tone cloning part has an ONNX model?
When I run the demo code directly, I get the error RuntimeError: CUDA failed with error out of memory. How much VRAM is required at a minimum? Or can I use the ONNX format to start the service?
Hi all, I have built the feature to export to ONNX. If anyone interested in, please contact me or check it here:
https://github.com/thanhnew2001/OpenVoice_ONNX