Turn a text or Markdown script into a natural TTS voiceover plus SRT/WebVTT captions timed to the spoken words and a per-word JSON timeline. Optional one-step video mux with ffmpeg. Chinese, English, Japanese. Free, MIT, pip install.