A voice assistant for your computer, powered by Porcupine, Whisper, ChatGPT and Azure TTS.
You need to have a few environmental variables set up before being able to run AIssistant.
PCP_ACCESS_KEY: PicoVoice access key, used for Porcupine for wake word detection (https://console.picovoice.ai/)
OPENAI_KEY: OpenAI api key, used for Whisper voice transcription and ChatGPT (https://platform.openai.com/)
AZURE_SPEECH_KEY and AZURE_SPEECH_REGION: Used for Azure neural TTS (https://azure.microsoft.com/)
After that follow these steps:
- Create a new venv:
python -m venv env - Activate the venv:
.\env\scripts\activate - Install the prerequisites:
pip install -r requirements.txt - Start the assistant:
python assistant.py
To run the assistant, just run the following commands:
- Activate the venv:
.\env\scripts\activate - Start the assistant:
python assistant.py
Thanks to dland on freesound.org for the assistant's activation cues: hint.wav