CaptionCraft guide
How to create automatic captions in Canva
Automatic captions turn spoken audio into editable on-screen text. The fastest workflow is not simply pressing generate: it combines clean source audio, a focused text review, readable styling, and one final preview before export.
Updated July 23, 2026 · 6 min read
Prepare the video before you transcribe
Start with the final cut of your video whenever possible. If you change the timing after generating captions, you may need to review the caption sequence again.
Clear speech produces a cleaner first draft. Reduce loud background music, avoid overlapping speakers when you can, and listen once for sections where names or specialist terms may need manual attention.
- Use the final or near-final video edit.
- Check that dialogue is louder than background audio.
- Write down names, brands, and technical terms you expect to correct.
Generate captions with CaptionCraft
CaptionCraft works directly inside Canva, so the video, transcript, styling, and preview stay in the same creative workflow.
- 1
Open the Canva design
Open the design that contains the video you want to caption, then launch CaptionCraft from Canva Apps.
- 2
Choose the video
Select the video you want to caption and start the automatic transcription.
- 3
Review the transcript
Read the full result once. Correct names, punctuation, numbers, and any words affected by music or accents.
- 4
Choose a template
Start from a caption template, then adjust its position, size, text properties, and effects to fit the design.
- 5
Preview and export
Watch the complete video with captions enabled. Check every scene change and export only after the text remains readable from beginning to end.
Edit for meaning, not only spelling
A correct transcript can still be difficult to read on screen. Keep each caption focused on one short idea, place punctuation where a viewer naturally pauses, and remove filler words when they do not change the meaning.
Pay special attention to the opening seconds. Viewers decide quickly whether to keep watching, so the first caption should be immediate, accurate, and easy to scan.
Run a final mobile preview
Most social video is watched on a phone. Preview at a small size and confirm that the captions do not collide with platform controls, logos, faces, or important visual details.
If the text feels crowded, shorten the caption before shrinking the font. Readability should win over fitting more words on screen.
Create captions directly in Canva
Generate, edit, style, preview, and export video captions with CaptionCraft.
Install CaptionCraft in Canva