girish-indurkar/VaaniShield ? reverse-engineered prompt

Build me a web app that helps spot voice cloning scams from uploaded audio.

I want a simple frontend where someone can drop in a voice recording and quickly see whether it sounds real or suspicious, along with a clear risk score, the transcript, and a few plain English reasons for the result. It should work with English, Hindi, and Marathi audio, and flag risky scam style phrases in the conversation too. Make the interface feel clean and easy to use, with a nice results view that shows the analysis clearly.

Use the backend to expose one upload endpoint that takes an audio file and language, runs the voice authenticity check, does transcription, and returns the combined result. If you need to look up current docs online for any library details, go ahead. Please make the whole thing runnable locally and keep the code organized so it’s easy to understand and test with sample audio.