The subtitle-to-text converter pulls just the words out of an SRT or VTT file and gives you a clean, readable transcript. It throws away the sequence numbers, timecodes and markup, and lets you collapse the result into single lines or a flowing paragraph — handy for reading, archiving, or pasting subtitled-video dialogue into a document.
How it works
The file is split into cues. For each block the tool skips the WebVTT header and any NOTE or STYLE
sections, skips the optional numeric index or cue identifier, skips the start --\> end timing
line, and keeps the caption text that follows. That text is then processed according to your
options:
- Strip formatting removes HTML-style tags and ASS-style override blocks.
- Collapse to one line joins a cue’s wrapped lines into a single space-separated line.
- Paragraph mode concatenates every cue into one flowing block of prose.
- Remove duplicates drops a cue identical to the previous one, which cleans up rolling captions that repeat lines as they scroll.
A live word count shows how much dialogue the transcript contains.
Example and notes
Three SRT cues such as \<i\>Welcome\</i\> to the documentary., Today we explore the deep ocean
and a repeated Welcome to the documentary. become, with tags stripped, duplicates removed and
paragraph mode on, a single clean sentence flow rather than three timed blocks.
Note: de-duplication only removes a cue that exactly matches the one before it, so distinct lines that happen to share words are preserved. Everything runs locally in your browser — no upload.