Mshavri/SoundPixel-AI — reverse-engineered prompt

Build me a simple Python app called SoundPixel AI.

I want to be able to put audio or signal files into an Inputs folder, run the app, and have it analyze them, improve or enhance the results using the AI and model pieces that are already in the project, then save clear outputs and reports in the Outputs and Reports folders. Make the app easy to run for a normal person, with a basic interface or local API if that is what fits the existing code best.

Please wire together the existing analyzer, signal core, AI enhancer, models, and gateway files so the project works end to end. Add any missing glue code, fix broken imports, create a clean start command, and make sure requirements are complete. Keep the design simple and reliable. If something is unclear from the empty README, use sensible defaults and document how to run it. Look up current docs online if you need to.