Not able to get diarization + ASR working

[pmarini-nc]

My goal is to have a meeting transcribed with diarization. I have the meeting recording in a file file.mp3.

The following call:

curl -X 'POST'   'https://<my-server>/v1/audio/diarization'   -H 'accept: application/json'   -H 'Content-Type: multipart/form-data'   -F 'model=vibevoice-crispasr'   -F 'file=@file.mp3;type=audio/mpeg'   -F 'include_text=true' -H "Authorization: Bearer <MYTOKEN>"

returns:

{"error":{"code":500,"message":"rpc error: code = Unknown desc = unimplemented","type":""}}

vibevoice-crispasr is running:

Configuration file of the model:

backend: cpu-crispasr
known_usecases:
    - transcript
name: vibevoice-crispasr
parameters:
    model: vibevoice-asr-q4_k.gguf

I'm on LocalAI 4.5.0. I've went through https://localai.io/features/audio-diarization/index.html, there are some aspects I don't fully understand and it would be great to have an end-to-end tutorial to have transcript+diarization tutorial.

Thanks!