git clone https://github.com/BasedHardware/omi.git cd sdks/python ### 2. Set up Python environment python3 -m venv venv source venv/bin/activate ### 3. Install dependencies pip install -r requirements.txt ### βοΈ 4. Configuration Set your Deepgram API key # Get a free API key from https://deepgram.com, then run: export DEEPGRAM_API_KEY=your_actual_deepgram_key 5. Find Omiβs MAC Address To connect to your specific Omi device, you need its Bluetooth MAC address. π Scan for Nearby Bluetooth Devices Run this command: python -c "from omi.bluetooth import print_devices; print_devices()" Look for a device named Omi, like this: 0. Omi [7F52EC55-50C9-D1B9-E8D7-19B83217C97D] Copy the MAC address inside the brackets and paste it into main.py: python Copy code OMI_MAC = "7F52EC55-50C9-D1B9-E8D7-19B83217C97D" ### 6. Confirm Omi Audio Characteristic (Optional) You can verify the audio characteristic UUID by running: python omi/discover_characteristics.py You should see something like: Characteristic UUID: 19B10001-E8F2-537E-4F6C-D104768A1214 This UUID is already used in the code by default. π Run the SDK Once configured, start the main script: python main.py β You should see: Connected to Deepgram WebSocket Connected to [Omi MAC Address] DATA FROM OMI: ... Transcript: hello world ### π§© Troubleshooting β Make sure Omi is powered on and near your computer β You must be using Python 3.10+ β If you see extra_headers error from websockets, fix it with: pip install websockets==11.0.3 ### πͺͺ License MIT License β this is an unofficial SDK built by the community, not affiliated with Omi. ### π Credits Built by [Your Name] using Omi hardware and Deepgramβs transcription engine.