> Project Overview: The mobile app should enable several types of user input: - Keyboard input: Fill out specific form fields - Voice input: Convert speech to text - Prompt input: Type or speak prompts for AI generation - Camera input: Capture and extract text using OCR Two processing approaches will be explored: - On-device (Offline mode): Perform OCR and simple generation locally without internet connection. - Cloud processing (Online mode): Send input data to a cloud backend for AI-based processing (e.g., transcription or image generation). The app must not contain or expose any cloud API keys, so a secure proxy or token-based access should be implemented via Azure Functions and API Gateway. Technical Requirements: - Framework: .NET MAUI (C#) – Android focus - OCR: Google ML Kit - Speech Recognition: Google Speech API (or Azure Speech Service alternative) - Cloud Backend: Azure Functions + API Gateway - AI Processing: Integration with a LLM or image generation API (e.g., OpenAI, Stability AI) - Simple UI for testing (design can be minimal; focus is functionality) Expected Deliverables: - Working MAUI Android app integrating OCR, speech input, and cloud AI - Secure connection to cloud backend (no hard-coded keys) - Simple and stable demo suitable for internal technical evaluation - Source code structured for easy extension in the PoC phase Use Case Examples: Shipping Label Reader: Capture an existing label with OCR or fill it by voice input. Idea Visualization: Generate a simple low-resolution image from a text or voice prompt.