Subtitle Generator — Speech-Assisted SRT, On-Device

Cueing your first SRT file to the second

The Subtitle Generator creates a timed .srt caption file from live speech recognition while your video plays out loud, all on your own device. On a 90-second product demo I hit Start listening, played the clip, and each spoken phrase landed as a numbered cue — a line recognized at the 4-second mark became 00:00:01,500 --> 00:00:04,000, since the tool gives each phrase the previous 2.5 seconds as its window. I fixed two misheard product names in the text boxes, hit Export .srt, and the file dropped straight into my editor's caption track.

Treat the recognizer as a first draft, not a stenographer. Brand names, jargon and crosstalk get mangled, so budget a proofreading pass — my 90-second demo needed six small edits before it shipped. In a noisy room, or with music under the dialogue, recognition degrades fast; the + Manual line button lets you add cues by hand with your own start and end times, which is also the fallback for languages your system's speech pack doesn't cover. Cue times live in editable number boxes, so a line that runs long is a two-second fix, not a re-record.