Mubumbutu/Audiobook-and-Dubbing-Generator ? reverse-engineered prompt

Build me a Windows desktop app for AI text to speech that can turn subtitle files, plain text, and ebooks into audio.

I want three main sections, one for subtitle fragments, one for ebooks, and one for quick text to speech. In the subtitle section, the app should load SRT, TXT, and ASS files, split them into chunks, generate audio for each line, and let me export the result with the original video audio mixed in. It should also support multi speaker dubbing by detecting speakers in a video, letting me assign voices, and then generating the spoken lines with the right voice.

In the ebook section, it should open common ebook formats, generate a full audiobook, and let me save and resume sessions. I also want export options like WAV, FLAC, MP3, OGG, and OPUS.

The app should support multiple local TTS backends, voice cloning from a reference WAV, and easy setup through installer and launch scripts. If you need to check current docs for any model or library, feel free to look them up online.