code-with-antonio/resonance — reverse-engineered prompt

Build me a polished open source style voice app that feels like a simple ElevenLabs alternative. I want users to sign in, create or join a team, and then turn text into speech with controls for things like creativity, variety, expression, and flow. They should be able to pick from a library of built in voices, upload or record a short voice sample to clone a voice, generate audio, listen in a nice waveform player, seek, pause, replay, and download the result.

I also want a clean dashboard that works well on mobile and desktop, plus a history page where people can revisit past generations with the voice details saved. Please include usage based billing so teams can be charged for characters generated and for creating cloned voices. It should be ready for self hosting with auth, Postgres, object storage for audio files, and a hosted GPU service for the speech engine. If anything is unclear, look up the current docs online and wire up sensible defaults.