Captions & transcripts
Captions & transcripts
Open a recording, click Generate captions, and get subtitles and a transcript next to the video. The speech recognition runs on your PC, so nothing is uploaded.
Windows 10/11 (x64) · Free · No account
The .srt and .txt are saved next to the video, named after it.

Use cases
When you'll reach for it — Captions & transcripts
Real situations, not a feature list.
- Add subtitles to a training video so people can watch it on mute.
- Get a written transcript of a recorded walkthrough to paste into documentation.
- Make a video easier to follow for viewers who are deaf or hard of hearing.
- Burn captions into a copy for a place that won't take a separate caption file.
What it does
Everything Captions & transcripts handles
Built into ScreenTray, free, and nothing you capture is uploaded.
Standard .srt captions
A caption file that video players and YouTube read, word for word with what was said.
A plain-text transcript
A .txt next to the video. Tick Clean up the transcript to remove fillers from the .txt only.
Burn into a copy
Burn captions into a copy draws them into the picture and saves a new video. The original is unchanged.
Nothing is uploaded
Whisper speech recognition runs on your PC, using the most accurate speech model you've downloaded.
Names spelled right
Your dictation dictionary helps captions spell the names and terms you use.
Runs in the background
Captions run in the Job Queue with progress, so you can close the preview and keep working.
How it works
Using it, step by step
- 1
Open a video
Double-click a recording in the library, or click Open on the card after you stop recording.
- 2
Generate captions
In Captions, click Generate captions. Optionally tick Clean up the transcript first.
- 3
Use the files
The .srt and .txt land next to the video. Show in folder opens them, or burn the captions into a copy.
Under the hood
How it actually works
What it uses, named plainly.
- Speech engine
- Whisper (whisper.cpp), on your PC
- Model
- The most accurate one you've downloaded; Base (about 150 MB) if you have none
- Output
- <video>.srt and <video>.txt next to the video; burned copy saved as … (captions).mp4
- Language
- The language set on the Dictation page, or detected automatically
- Platform
- Windows 10/11 (x64); speech recognition needs a CPU with AVX2
Captions are as accurate as the speech model: names, jargon and noisy audio can come out wrong. Pick a larger model, set the language and add names to your dictionary, then Generate again. A video with no speech or no audio track gets no captions.
Also in ScreenTray
The rest of the capture kit
Questions
Captions & transcripts, answered
Get ScreenTray — Captions & transcripts included
Captions & transcripts is part of ScreenTray, free for Windows 10 and 11, with the rest of the capture kit alongside it.
Windows 10/11 (x64) · Free · No account