video-voiceover

Synthesize Chinese narration audio (TTS voiceover) from a timestamped narration.json. Use to turn a written narration script into per-segment speech audio, with MiMo TTS (mimo-v2.5-tts), dynamic speed fitting, and loudness handling. Part of the video-recap bundle: consumes narration.json (output-timeline time), produces tts_segments + tts_meta.json. In the legacy direct-cut path, narration_mapped.json may be passed explicitly instead. 触发词: 配音, 语音合成, TTS, 解说配音, voiceover, text to speech, 旁白配音.