WotPan AIPrivate by design
Speech recognition runs on the phone with OpenAI's Whisper (via whisper.cpp). Your audio is never uploaded.
✦ Runs on your phone
AI Captions Generator for Reels, Shorts & Vlogs
WotPan AI listens to the speech in your video and turns it into animated, word-by-word captions. In Hindi, Hinglish, English, Arabic and 12 more languages. It all happens on your phone, so your audio never leaves it.
Free app for Android · iPhone app coming soon
Auto captions
How to add auto captions to a video
Three steps in the WotPan AI app. No typing, no timing each line by hand.
- 1
Open your video
Start a project with your clips, then tap Captions in the editor toolbar (or Text → Auto captions).
- 2
Pick the spoken language
Choose the language people speak in the video, or leave it on Auto detect. Pick a caption style and how many words show at once.
- 3
Generate and fine-tune
Tap Generate. Captions land on your timeline as text you can edit, restyle and animate, then export your video.
Captions in 16 languages, including Hindi and Hinglish
Made for creators who switch languages mid-sentence. Pick a language, or let the app detect it.
- Auto detect
- Hinglish (Roman Hindi)
- Hindi (हिन्दी)
- Urdu (اردو)
- English
- Spanish
- Arabic (العربية)
- French
- German
- Portuguese
- Indonesian
- Turkish
- Russian
- Japanese
- Korean
- Chinese
- Italian
Hindi gives you Devanagari script; Hinglish writes the same speech in Roman letters, the way most people type it. On Auto detect, Hindi or Urdu speech is written in Hinglish.
Why creators use WotPan AI for captions
Works offline after setup
The speech model downloads once, the first time you use it. After that, captions don't need the internet.
Fast or accurate
Choose the Fast model for quick drafts or Accurate for tricky audio. Hindi and Hinglish use a model tuned for Hindi.
Four caption styles
Clean white with an outline, bold yellow, big uppercase with a shadow, or text on a dark box, ready for Reels and Shorts.
Words per caption
Show one word at a time for punchy edits, or up to ten for calmer talking videos.
Edit every line
Captions are normal text clips: fix a word, restyle or animate them, split, duplicate or delete them.
Frequently asked questions
Is the AI captions generator free?
Yes. Auto captions are part of the WotPan AI app, which is free to download on Google Play. The app shows ads. Exporting a video asks you to sign in with Google.
Does WotPan AI upload my video or audio to make captions?
No. Speech is recognised on your phone with OpenAI's Whisper model (through whisper.cpp). Your audio never leaves the phone. The internet is only needed once, to download the speech model the first time you use it.
Can it make Hindi or Hinglish captions?
Yes. Pick Hindi for captions in Devanagari script, or Hinglish (Roman Hindi) for Hindi speech written in English letters. For Hindi and Hinglish the app uses a speech model tuned for Hindi. Urdu is supported too, in its own script.
Can I edit the captions after they are generated?
Yes. Captions are added to your timeline as normal text clips, so you can fix a word, change the style or animation, split, duplicate or delete them like any other text.
Does the web editor make automatic captions?
Not yet. Automatic captions from speech are in the WotPan AI Android app. The web editor has a Captions panel where you type each line and it is placed on the timeline for you.
How big is the speech model download?
It downloads once, the first time you use it: about 32 MB for the Fast model, 60 MB for the Accurate model, or 148 MB for the Hindi model.
Caption your next video in a minute
Get WotPan AI free on Android. Auto captions, templates, background remover and up to 4K export in one app.
Get it on Google Play- 16 languages plus Auto detect
- Speech recognised on your phone
- Captions you can edit and restyle
- Free to download