jaidh01/Speech-to-Image ? reverse-engineered prompt

Build me a simple Streamlit app that lets a user speak a prompt, turns that speech into text, and then uses that text to generate an image.

I want it to feel easy and fun to use, with a clean interface where someone can press a button, talk into their microphone, see the transcribed prompt on screen, and then generate a matching visual. If the speech is unclear or nothing is detected, please show a helpful message instead of failing silently.

Keep it as one main Python app file if possible, and make it straightforward to run from the command line with Streamlit. If you need current docs or the best way to wire up the speech and image parts, look them up online. Also make sure the app is polished enough that I can clone it and try it right away without a lot of setup confusion.