Convert VTT to Plain Text

They are skipped. The WebVTT spec forbids the cue arrow inside them, which is exactly how the parser tells them apart from real cues — nothing is guessed at.

Choose DOCX or PDF instead of TXT; both put a timestamp in front of each line. Plain text drops them by design, because the usual reason for wanting .txt is to paste the words somewhere else.

YouTube auto-captions roll each line into the next cue so the text scrolls on screen, so a straight export shows every line twice. Undoing that reliably needs the original audio — transcribing the video with ScribeToAny gives clean, non-overlapping segments instead.