Automatic captions and motion titles
octuedit adds styled captions to your videos automatically. It transcribes speech on your Mac with whisper, timed to each word, then lays captions out in one of 18 styles, from clean subtitles to bold word-by-word highlights, keeping them clear of the Reels, TikTok and Shorts buttons. You can ask for changes in plain words, edit any caption by hand, and add motion titles, lower thirds and title cards. Captions are burned into your export or saved as SRT or WebVTT files, and nothing is uploaded for transcription.
Updated 27 October 2026 · By Shivank Goura, co-founder of octuedit
What can you ask for with captions and titles?
Type the edit you want in plain words and your own Claude Code or Codex carries it out on a real timeline, which you can still adjust by hand. These requests work well for captions and titles; each one shows what octuedit does with it.
- “Add bold captions that highlight each word as it's said”Transcribes on your Mac and adds word-by-word highlighted captions.
- “Use the editorial caption style and put them behind me”Switches the style and places the text behind you using a background mask.
- “Add a title card at the start and a lower third with my name”Adds a motion title and a lower third you can restyle.
- “Spell Octupie correctly everywhere”Uses your custom vocabulary so names and brands come out right.
Can I add styled word by word captions to my videos automatically?
Yes. octuedit, an AI video editor for Mac, transcribes your footage on your Mac with the whisper large-v3-turbo model, which times every word, then builds captions in any of 18 styles. Captions can show the whole line, highlight each word as it's spoken, reveal words one at a time or show a single word, and one stress word per card can be emphasised. They stay clear of the Reels, TikTok and Shorts buttons and can sit behind the speaker. You ask in plain words, such as "bold captions that highlight each word", and fix any line by hand; a locked line keeps your text when captions rebuild. Up to 60 names and brands can go in a custom vocabulary so they're spelled your way. Captions are burned into MP4, MOV or WebM exports or saved as SRT or WebVTT files, and nothing is uploaded for transcription. octuedit costs $99 once.
How does captions work in octuedit?
When you import footage, octuedit transcribes it on your Mac with the whisper large-v3-turbo model, which gives each word its own timing. It detects the spoken language automatically, handles Chinese, Japanese and Korean text, and writes Hindi as readable Hinglish rather than translating it. You can add up to 60 names and brands to a custom vocabulary so they are spelled your way.
Captions are built from those word timings in the style you ask for. Each caption card can show the whole line, highlight the spoken word karaoke style, reveal words as they are spoken or show one word at a time, and one stress word per card can be emphasised. Captions avoid the platform interface automatically and can sit behind the speaker. Every card stays editable: fix a word by hand and lock it so a rebuild keeps your text.
| Capability | In octuedit |
|---|---|
| Caption styles | 18 presets including Clean, Highlight, Stacked, Word Pop, Boxed, Beast Mode, Editorial, Subtitle, Documentary, Typewriter and News |
| Animation | Whole line, karaoke highlight of the spoken word, reveal as spoken, or one word at a time |
| Emphasis | One stress word per card, such as a number or a negation, set apart in colour or a second typeface |
| Placement | Kept inside the Reels, TikTok and Shorts safe area; can sit behind the subject |
| Transcription | On your Mac with whisper (large-v3-turbo), word-level timing, automatic language detection, custom vocabulary up to 60 terms |
| Translation | Captions can be translated by your Claude agent, keeping their timing |
| Titles | Title, subtitle, lower third, caption box, serif quote and punchy text styles, with 12 motion presets such as fade up, pop, typewriter and word by word |
| Output | Burned into MP4, MOV and WebM exports, or saved as SRT or WebVTT files |
What should you know about captions in octuedit?
- Bundled fonts include Bebas Neue, DM Serif Display, Lilita One and Abril Fatface, plus your Mac's own fonts.
- Transcription never leaves your Mac; translation uses your own Claude account.
- You can edit and lock any caption so later changes don't overwrite it.
What can't octuedit do with captions?
- Burned-in caption styling stays in the video; SRT and VTT files carry the words and timing only.
- Accuracy depends on audio quality and accents, as with any speech model; check names and numbers before publishing.
- Caption styling doesn't carry over in project exports to Premiere Pro, Final Cut Pro or DaVinci Resolve; export SRT for those.
Questions
Does octuedit add captions automatically?
Yes. It transcribes your footage on your Mac and adds timed captions in the style you choose. Ask for a style in plain words, such as "bold captions that highlight each word", and adjust any line by hand.
Can octuedit do word by word captions like CapCut and Submagic?
Yes. Captions can highlight each word as it is spoken, reveal words one at a time, or show a single word at a time, across 18 caption styles.
Which languages does octuedit transcribe?
It uses the whisper speech model with automatic language detection, so it is not limited to English. It has specific handling for Chinese, Japanese and Korean, and writes Hindi as romanized Hinglish.
Can I export captions as SRT or VTT?
Yes. octuedit exports SRT and WebVTT caption files from the same timing, or burns captions into MP4, MOV and WebM exports.