How to turn a transcript into SRT subtitles

Turning a transcript into SRT subtitles requires one thing above all: timestamps. If your transcript carries them — and every export from Rev, Otter, Descript, or Whisper can — the conversion is mechanical: each timestamped passage becomes a numbered SRT cue with millisecond start and end times, split into caption-sized lines. If your transcript is bare text with no timing, no honest tool can convert it, because subtitle timing has to come from somewhere real.

Have a timestamped transcript now? Drop it on the free TXT to SRT converter — Rev, Otter, Word-style, and Avid caption exports all work, speakers included. Whisper or other JSON output goes through the JSON to SRT converter instead.

What your transcript needs to look like

Any of these shapes converts cleanly:

• Rev-style, either the tabular export (Timestamp / Speaker / Transcript columns) or the classic delivery format:

Speaker 1 (00:02):
Welcome back to the show.

• Otter-style, speaker-first with a timestamp on the name line:

Sarah Chen  0:02
Welcome back to the show, everyone.

• Word-style / generic, bracketed or leading timestamps:

[00:00:02] Sarah: Welcome back to the show.

• Whisper and other AI output as JSON, where each segment carries start and end times in seconds — use the JSON to SRT converter for these.

The common thread: a time anchor attached to each speaker turn or passage. Speaker names, when present, survive into the SRT as labels (Sarah: Welcome back…).

How the conversion works

1. Export the transcript from your service with timestamps enabled (and speaker names, if you want them) — Otter, Rev, and Descript all have the checkbox in their export options, and Rev's guide to transcript file formats is a good overview of what each export contains. This is the step everyone misses — a bare-text export cannot be timed after the fact.
2. Drop the file on the TXT to SRT converter.
3. The converter reads each timestamp, splits long passages into readable caption-length cues between the anchors, and stamps every cue with standard SRT millisecond timing.
4. Download the .srt — or a single ZIP if you converted a batch.

The output is standards-compliant SubRip (SRT) — the plain-text subtitle format that uploads to YouTube and Vimeo, loads into Premiere, Resolve, and Final Cut, and opens in any subtitle editor.

The timing question, answered honestly

Transcript timestamps mark where a paragraph or speaker turn starts — not where every sentence begins and ends. The converter times each cue from the anchors your file actually contains: a cue starts at its timestamp and runs until just before the next one, with long passages split into multiple cues across that span and a readable duration estimated for the final cue. That produces well-synced subtitles from paragraph-level timestamps — but the denser your timestamps, the tighter the sync. An Otter export with a timestamp per speaker turn syncs noticeably better than a lecture transcript with one timestamp per five minutes, and Whisper JSON (a timestamp per segment) is tighter still.

Where the SRT goes next

• YouTube / Vimeo / players: upload the .srt directly.
• Premiere Pro: SRT imports as captions directly; to use it in the transcript panel for Text-Based Editing instead, take the transcript JSON route.
• DaVinci Resolve: import the SRT onto a subtitle track.
• Avid Media Composer: convert once more with SRT to Avid TXT and import via SubCap.
• Word or a spreadsheet for review and translation: SRT to Word or SRT to CSV.

FAQ

Can I convert a transcript with no timestamps into SRT? No — and be suspicious of any tool that claims otherwise. SRT cues need real start and end times, and timing has to come from somewhere. Re-export from your transcription service with timestamps switched on; every major service supports it.

Which transcript formats convert directly? Rev (tabular and the parenthesized delivery format), Otter speaker-first exports, Word-style bracketed timestamps, Avid caption TXT, and generic timestamped text. Whisper and other AI JSON output converts via the JSON to SRT converter.

Do speaker names end up in the subtitles? Yes, when the export includes them — each speaker turn becomes a labeled cue, like "Sarah: Welcome back." If you don't want labels, export without speaker names.

How accurate is the subtitle timing? As accurate as the timestamps in your file. Paragraph-level timestamps give solid sync with cue boundaries estimated inside each passage; per-turn or per-segment timestamps (Otter, Whisper) give tight sync throughout.

My transcript is a Word document, not TXT — does it work? Save it as plain text first, or if it's a timed transcript destined for Premiere, the DOC to Premiere JSON converter reads .docx directly. For SRT output, plain-text export with timestamps intact is the reliable path.

What's the difference between this and pasting my transcript into a subtitle editor? About an hour of retyping timestamps. A subtitle editor makes you place every cue by hand; this reads the timing your transcript already carries and builds the cues in seconds.