xtts-webui

Why we included this project

This web UI puts Coqui's XTTS v2 behind a browser interface, so you can synthesize realistic speech without scripting around the model. It also handles batch processing for dubbing large sets of files, keeps the original voice when translating audio, and can route results through RVC, OpenVoice, or Resemble Enhance to clean them up. Fine-tuning is bundled into the same interface, letting you adapt the model to a custom voice and generate with it right away. A portable Windows build runs on an NVIDIA card with 6 GB of video memory, which removes most of the setup pain. Teams doing voiceover, video dubbing, or voice-cloning experiments will find the all-in-one workflow the main draw.