I’m building a compact Raspberry Pi-based device that combines a camera, microphone, and speaker, and I need two tightly-linked pieces of software: • Embedded firmware that lives on the Pi and takes charge of the hardware. While the camera and speaker must be initialised, the heart of the firmware is robust microphone and audio processing—clean capture, basic filtering, and a reliable audio stream exposed to the higher-level app. Low-level control should be in C/C++ or performant Python with GPIO/I²S handled properly. • A companion application that runs on the same Pi and focuses on real-time monitoring. This layer adds AI capabilities: – Image recognition from the camera feed – Voice commands parsed from the mic stream – Automated responses through the speaker and display alerts I’d like clean, well-commented source code, a simple way to deploy (Docker or a scripted install), and a short README that shows me how to start the firmware service and the monitoring app together. If you lean on familiar tools—OpenCV, TensorFlow Lite, or PyTorch for the vision side, and something like Vosk or SpeechRecognition for voice—it will speed things up, but feel free to suggest better options as long as latency stays low. Testing will be done on my own Pi 4 with the official camera module and a USB microphone, so please keep hardware assumptions minimal. Once I can boot the Pi, see the live feed in the app, talk to it, and watch it react instantly, the job is complete.