Custom Hugging Face model plugin

I have a fine tuned model for transcribing Tibetan audio which has low CER. I have hosted the model in Hugging face.
Can i have option to plugin custom model.

The issue does not name a file, test, or entry point. First determine how transcribee currently selects and loads transcription models, then clarify the plugin interface and supported Hugging Face model requirements. Done means a custom hosted model can be selected and used for Tibetan transcription with coverage for the integration behavior.

Written by the indexing model from the issue text.