Captions

Ask for captions and Montage writes the words you say on screen, timed to your voice, in the font, colour and place you describe.

Written by Ben (Founder) · Updated November 5, 2026

How Montage knows the words

When a footage or a music with words arrives, Montage listens to it and writes down every word with the moment it is said. That is what the captions follow, word by word.

Add captions, step by step

Open your video and, in Ask, write what you want: « Add captions, big and white, at the bottom ».
Name a font if you have one in mind: any Google Fonts family (« in Anton »), or a font file you added with + then Add footage.
Or pick a style from the Library (+, then From the Library): the captions follow it.
Watch, then ask again or comment on a moment: « @0:12 this word is wrong ».

What to ask

You want	Ask
One word at a time	« Show the captions one word at a time »
Another look	« Yellow, with a black outline »
Another place	« Move the captions above my chin »
A word fixed	« At 0:12 it’s “Montage”, not “montage” »
The same captions in other videos	« Save these captions as a component »

Arabic, Hebrew and Persian are written right to left.

Questions

Which languages work?

Can I remove the captions later?

Yes. Ask « remove the captions », or go back to an earlier line of the History.