A powerful 3B-parameter, LLM-based Reinforcement Learning audio edit model excels at editing emotion, speaking style, and paralinguistics, and features robust zero-shot text-to-speech

[CVPR 2025] Official source codes for the paper: EmoDubber: Towards High Quality and Emotion Controllable Movie Dubbing.

MLX implementation of IndexTTS (v1.5 & v2.0) — high-quality text-to-speech with voice cloning and emotion control, optimized for Apple Silicon

🎙️ Beautiful FastAPI WebUI for IndexTTS2 on Apple Silicon — dark theme, 8 emotions, custom reference audio, one-click NPZ speaker export

NL interface surpassing GPT limits: emotion control, parallel commands, quantified inputs like "remove 30% emotion".

AI Text-To-Speech is a versatile desktop application that converts text into natural-sounding speech using leading AI providers and local models. It integrates ElevenLabs, Azure Cognitive Services, and Google Cloud TTS for cloud synthesis, plus Bark and Tortoise-TTS for fully offline generation.