using a custom model

[jeyhunbay]

I have a fine-tuned model that converted it to faster-whisper format using the instructions in the faster-whisper repo. I can load that model with faster-whisper and it works. How to add that model to the server?
I tried changing the env variable WHISPER__MODEL to point to my model's path, but the model is not being picked up.
Any suggestions?
Thanks!

[jeyhunbay]

for now I just patched list_whisper_models in faster_whisper_server/hf_utils.py to inject my custom model to the list of available models. Probably not the most elegant solution, but it at least it works. But still looking for a better way of doing this.

[fedirz]

If you set the HF_TOKEN environment variable on HuggingFace, your private model should appear in the model list. However, loading a local model (not on HuggingFace) is currently unsupported. I plan to support this, but I am unsure when that will happen.

Thanks for using the project and creating an issue!