StyleTTS2 is a text-to-speech model that generates high-quality speech with controllable style - View it on GitHub