auracletech/piai

Use Pi AI hands free ✒

nice.mp4

Pi AI Hands Free

a python AI Voice Interpreter for Pi AI

Installation

Follow these instructions in order to get the application running.

1. Install requirements

• git
• FFmpeg
• Python 3.11

2. Clone this repository to your local machine using the following command

git clone https://github.com/AuracleTech/piai.git

3. Navigate to the project directory

cd piai

4. Install OpenAI Whisper

pip install -U openai-whisper

5. Install pipreqs to generates the dependencies

pip install pipreqs

6. Generate the dependencies file

pipreqs . --force

7. Install the dependencies

pip install -r requirements.txt

8. Launch the app

python src/main.py

9. Have a chat with your new voice assistant 🌟

Development

The application uses 3 threads

1. Recorder.py records vocal inputs to recordings folder.
2. Transcriber.py transcribes the files and sends the transcript to the interpreter.
3. Interpreter.py transmits what you said to Pi.

OpenAI

You can use GPT to determine if Pi should respond to the vocal input, to use it follow these steps.

1. Get an OpenAI API Key

2. Create a .env file in the root directory and add the following variable

OPENAI_API_KEY=your-openai-api-key-here

3. Uncomment the code in interpreter.py and/or modify it to your liking.

Help

Feel free to open an issue if you have any questions, suggestions or issues.