DanielSWolf/rhubarb-lip-sync — reverse-engineered prompt

Reverse engineered prompt

GitHub

Build me a command line lip sync tool for 2D animation. I want to give it a voice recording, like a WAV or Ogg file, and have it automatically figure out the timing for mouth shapes so an animator or game can use it.

It should output a simple file with time stamps and mouth shape letters, with the basic shapes A through F and optional extra shapes G, H, and X for rest, F and V sounds, and long L sounds. Please support common export formats like TSV, XML, and JSON, and also a format that works for animation tools like Moho or OpenToonz if practical.

Make it usable from the terminal on Windows, macOS, and Linux, with clear options for choosing the speech recognizer, output file, export format, frame rate when needed, and whether to use extra mouth shapes. Good error messages matter. Use C++ if that fits best, and look up current docs online if you need to.

Want more depth? Deep Reverse