Subtitles help more people watch your video, including everyone who watches with the sound off.
Step by step
- Open the Subtitle Generator (SRT · VTT) and add your video file.
- Pick a model. For best accuracy choose small. The model downloads once the first time.
- When recognition finishes, play the video and fix the subtitle text and timing.
- Split long lines with "Split long lines" and nudge out-of-sync subtitles with the ±0.5 s buttons.
- Save as SRT or VTT.
Where to use an SRT file
- YouTube: after uploading your video, upload the file from the Subtitles menu.
- Video editors: bring it in with the subtitle/caption import feature.
- VTT: commonly used by web video players.
Good to know
- Automatic recognition gets proper names and jargon wrong. Always skim the result.
- Videos with loud music or songs come out less accurate.
- Videos with several speakers do not get speaker labels in subtitles.
Sources and date checked
Checked: October 3, 2026
- Wikipedia, SubRip — https://en.wikipedia.org/wiki/SubRip
This article is general information. Devices and services can change over time.
See the YouTube & podcast audio workflow for the full production order.
Frequently asked questions
What is the difference between SRT and VTT?
Both are timed subtitle text files. SRT is widely supported by YouTube and video editors; VTT is common in web video players.
Is my video uploaded to a server?
No. Recognition runs in your browser and the video stays on your device.
Why are some subtitles out of sync?
Automatic timing can drift slightly. Use the ±0.5 s buttons to shift lines while you watch.
Can it label who is speaking?
No. Subtitles are not split by speaker. For speaker labels, use the AI Meeting Notes tool.