This repository provides tools for rapid speaker adaptation of VITS (Variational Inference for Text-to-Speech) models, enabling voice cloning from short audio clips, long recordings, or videos within under an hour of fine-tuning. It supports English, Japanese, and Chinese TTS synthesis as well as many-to-many voice conversion between added characters and preset speakers in the model. Users can fine-tune locally or via Google Colab to create custom voice profiles.