/Orallama

Primary LanguagePythonGNU Affero General Public License v3.0AGPL-3.0

Orallama

Great thanks to the reference project of this work: apeatling/ollama-voice-mac

Installing and runningšŸ¦™

  1. Install Ollama.
  2. Download the a model, Mistral 7b is a actually a ood choice using the ollama pull mistral command.
  3. Download an OpenAI Whisper Model and place the .pt model file in a /whisper directory in the repo root folder.
  4. Make sure you have FFmpeg by running brew install ffmpeg.
  5. For Apple silicon support of the PyAudio library you'll need to install Homebrew and run brew install portaudio.
  6. Run poetry install to install all dependencies (pip install -r requirements.txt also works).
  7. Run python chat.py to start the assistant.