Generate a voiceover and export SRT cues based on the actual lengths of its audio lines and pauses.

Make the final voiceover first, then export the SRT alongside it. Each generated line becomes a caption cue timed to that line's audio.

1. Keep captions readable

Break long narration into shorter lines before generation. One line becomes one SRT cue.

2. Generate the final audio

After changing text or speed, regenerate the affected line. The caption timing should use the final clip, not an old take.

Download the full audio and SRT together. Open them in your video editor and retime the captions if you move or stretch any clips.

Make two short lines with a one-second pause after the first. The second SRT cue should begin after that gap in the exported recording.

A narration where sentence-level caption cues are a useful starting point.

Limit: The SRT is line-level, not word-level. It does not stay synchronized after audio edits in another app.