Strips the timings and turns the cues back into readable prose. The naive version of this — delete every timestamp and keep the lines — produces a column of six-word fragments that nobody can read, so this one reassembles them.
How the paragraphs are found
Consecutive cues are joined into one flowing paragraph, and a new paragraph starts wherever there is a real gap in the dialogue — 2.5 seconds by default, which corresponds closely to a scene change or a pause between speakers. Adjust it if the source is unusually fast or slow.
If you want the raw cue list instead — for a translation memory, or to diff against another version — switch the layout to one cue per line.
What is removed
<i> are all stripped. Speaker labels written into the text (- JOHN: Hello) are kept, because they are part of the text rather than part of the format.
Turn on the timestamp option if you want each line prefixed with its start time — useful for referring back to a specific moment in a long recording.
What a transcript is good for
Other names for this
Also searched as “srt to transcript”, “subtitle to text”, “extract text from srt”.
Questions
- Why are my sentences split across lines?
- They were split across cues in the source. Paragraph layout rejoins them; one-cue-per-line deliberately does not.
- Can I keep the timestamps?
- Yes — turn on “Prefix each line with its timestamp” and each line starts with its start time.
- Are speaker names kept?
- Yes, when they are written into the cue text. They are content, not formatting.
- Is my subtitle file uploaded?
- No. The parsing and rewriting happen in this tab. Subtitle files often contain an unreleased script or a client’s content, which is a good reason not to send them to a stranger’s server. Turn your Wi-Fi off and the tool keeps working.
- Can I convert several files at once?
- Yes — drop as many as you like, or a whole folder. Each keeps its original name with the new extension, and you can download them individually or as a ZIP.