Drop your video in, pick a style, export. OpaCaptions does the hard parts for you: it listens, writes down every word with
its timing and puts the captions on your video, all on your own PC. No editing skills, no tutorials, nothing to set up first.
1window, nothing else to learn
3steps: drop, pick, export
~10 sfor a 30-second clip on an ordinary laptop
0editing skills needed
OpaCaptionsFileLicenceHelpYour AI, your PCPro plan · v0.15
Transcript
Detected: English · 28 words
Search what was said
Drop a video or audio fileMP4, MOV, MKV, MP3, WAV · any size
0%Transcribing on this PC · nothing uploaded
Get the speech model ready
Read the audio
Turn speech into words
Transcribing on this PC
0%
Just heard
0:00Your video never leaves this PC
StylesCustomiseOverlays
WORD POPBold Popglow upNeonClear & crispClassicquiet & calmMinimalOnce uponCinema
ExportMP4 video + SRT + VTTCtrl+E
00:00:00.001×
Done. Nothing left your PC.
launch-teaser.mp4Video with captions burned in
launch-teaser.srtSubtitles for YouTube & editors
launch-teaser.vttWeb subtitles
launch-teaser.txtSearchable transcript
Saved to Videos › OpaCaptions
launch-teaser.mp4184 MB · 0:15
Interface preview: OpaCaptions is in early access. A 30-second clip takes about 10 seconds on an ordinary laptop in Fast mode.
Who does what
You do three things. OpaCaptions does the rest.
If you can choose a filter, you can do this.
You
Drop your video inDrag it onto the window, or click Open.
Pick a styleHover over any style and it plays on your video before you choose.
ExportOne click saves the captioned video, ready to post.
Optional: click any word to fix it. Typing the right word is all it takes.
OpaCaptions, automatically
Recognises the language and writes down every word
Times each word to the moment it's spoken
Breaks the words into easy-to-read caption lines
Places the captions clear of the app buttons on vertical videos
Tells speakers apart and labels them (Pro)
Styles and animates the captions, then saves your video, subtitles and transcript
Each step in detail
Curious what's happening? Here's every step.
1
Drop in a file
Drag any video or audio file into the window: MP4, MOV, MKV, WEBM, MP3, WAV or M4A, of any size, up to 4 hours long. Vertical, landscape and audio-only all work.
The file is read straight from your drive. Nothing is copied anywhere else.
2
Transcribe
OpenAI's open-source Whisper model listens and writes down every word with its exact timing. Pick Fast, Balanced or Best accuracy, and the spoken language, or let it detect the language for you.
Whisper runs on your own processor. No graphics card, no cloud.
3
Style
Choose a caption style and see it on your video straight away; on Pro, tweak its font, colours, size and position. One switch turns any of the 17 languages into English captions.
The preview is live; nothing is rendered until you export.
4
Review
Read the transcript beside the video. Every word lights up as it's spoken, search finds any moment, and a click on any misheard word lets you fix it, on every plan.
Your transcript stays a file on your PC, not a record on someone's server.
5
Export
Get an MP4 with the captions burned in, ready for TikTok, Reels and Shorts, plus SRT and VTT subtitle files and a TXT transcript, saved to your folder.
Rendering uses ffmpeg on your PC. Upload counter: still 0 bytes.
Requirements
Nothing to set up first. Your PC probably has what it needs.
Install it, open it, drop in a video. The Fast speech model comes with the app, so your first video doesn't wait for a download.
Windows 10 or 1164-bit. A Mac version is on the roadmap.
8 GB of RAMAn ordinary laptop is enough.
No graphics cardEverything runs on the processor.
About 150 MB to 1.6 GBFor the speech model, downloaded once, depending on the accuracy you use.
Internet, barelyTo download the model, activate your licence and check for updates. Captioning itself works offline.
Speed and accuracy
Fast when you need it. Careful when it counts.
On the 8th-generation Intel Core i5 laptop we build on, a 24-second clip is captioned and rendered in about 9 seconds in Fast mode. Faster PCs simply finish sooner.
Mode
Best for
Model download
Fast
Clear speech, quick social clips
about 150 MB
Balanced
Most videos and podcasts
about 0.5 GB
Best
Accents, background noise and technical words
about 1.6 GB
Balanced and Best take longer than Fast, in exchange for catching more words in difficult audio.
Works offline
Switch on airplane mode. It keeps working.
After the one-time model download and licence activation, OpaCaptions needs no internet at all. That's the simplest proof that your footage never leaves your PC: unplug the network and caption a video. Read more in our privacy policy, or see what stays on your PC.