search

savbell / whisper-writer

💬📝 A small dictation app using OpenAI's Whisper speech recognition model.
GNU General Public License v3.0
244 stars 40 forks source link

LocalAI support for locally hosted Whisper Models #53

Closed danobot closed 1 month ago

danobot commented 1 month ago

It would be great to add instructions for the abilty to use the LocalAI API to make use of a locally hosted Whisper model on a more powerful machine (rather than using the Base or Tiny model on the machine running whiser-writer).

Audio to text: https://localai.io/features/audio-to-text/

savbell commented 1 month ago

Hi, thanks for your comment! You can use the LocalAI API by choosing "Use API" on the first tab in the Settings window and then entering in the local endpoint under "Base URL", e.g. http://localhost:8080/v1. Hope that helps! :)

danobot commented 1 month ago

Thank you, this worked perfectly. I had to edit the config.yaml manually to set model: ggml-faster-whisper-large-v3 and edit .env to give a random value for api key (its not used by localai but the value must be populated it seems)