README
This repository contains the quantized models and the scripts for converting ONNX models to FP16 format. These files are intended to be used with the original CosyVoice2 inference code.
Usage
To use the quantized models, you have two options:
Download and Rename the Files:
- Download the FP16 models.
- Rename the downloaded files to match the expected names in the original CosyVoice2 inference code.
Change Mentions in the Original Inference Code:
- Modify the original CosyVoice2 inference code to reference the new FP16 model filenames.
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
HF Inference API was unable to determine this model's library.