Add model cache path option to Docker run instructions in readme

[6a67]

No description provided.

[kanjieater]

Thanks for the PR. I didn't know where the models were being downloaded to, but it seems like the path you have listed is where the docker container downloads them to, is this correct? If that's the case, this is a good find, and I'll merge this.

[6a67]

Exactly. It seems the tiny and large Whisper models inside the container are located at:

/config/.cache/huggingface/hub/models--Systran--faster-whisper-tiny
/config/.cache/huggingface/hub/models--Systran--faster-whisper-large-v3

Meanwhile, the VAD files appear to be stored at:

/config/.cache/torch/hub/snakers4_silero-vad_master/src/silero_vad/data

Given this, I figured it would make sense to simply mount the entire .cache directory.