...
Guide users through the FlyCut Caption workflow: take a raw video or audio file, run Whisper-based speech-to-text to get timestamped subtitles, prune filler / unwanted segments by ticking subtitle lines, then export either clean subtitle files (SRT/JSON) or a re-cut video with only the kept segments (optionally with burnt-in captions). Use this skill when a user wants to remove "嗯/啊/废话" filler, tighten a talking-head video, or generate a subtitle file from raw footage without learning a full NLE.
Créé par: songlin she · GitHub