jamiepine/voicebox ? reverse-engineered prompt

Build me a desktop app for voice synthesis that runs completely locally on my computer. I want an open-source alternative to something like ElevenLabs, where I can clone a voice from a short audio clip and then generate new speech from text. It should be private, with all the models and data staying on my machine.

The app needs to support multiple languages and give me creative control. I want to be able to add audio effects like reverb, pitch shifting, or making it sound like it's on the radio. It would also be great if I could type special tags like [laugh] or [sigh] directly into the text to make the speech more expressive.

Please also include a simple timeline editor. This would let me arrange clips from different cloned voices to create conversations or podcast-style audio. The app should be high-performance and work on both Mac and Windows.