# Add Styled Captions to a Video (Add Captions) Source: [Add Styled Captions to a Video (Add Captions)]() Documentation index: [llms.txt]() Last updated: 2026-11-08 The **Add captions** Skill transcribes speech in your video, enables word-timed captions on a named editable timeline, and applies the exact built-in preset you picked in the style gallery. It keeps your original video, spoken words, audio, and aspect ratio intact without trimming the clip. You can run it from **Skills** in ChatCut Web or ChatCut Desktop, or ask a connected external agent or the Agent Plugin to load and run it on your project. ## When should I use Add Captions? Use it when you have a spoken-word video and want to preview and pick a specific built-in caption style up front, then have the Agent transcribe the speech, apply confirmed spelling fixes, and style the captions on an editable timeline in one step. Do not use it in these cases: - **You want to clean up retakes, filler words, or long pauses before captioning.** Use [Speech Cleanup](https://pr-824-preview.chatcut.dev/docs/speech-cleanup) or [Talking Head Edit](https://pr-824-preview.chatcut.dev/docs/talking-head-edit) first so you do not style speech you plan to cut. - **You want a bespoke animated caption or kinetic typography template built from scratch.** Ask the Agent for a custom animated caption template instead; see [Caption Styles](https://pr-824-preview.chatcut.dev/docs/caption-styles#can-the-agent-make-a-custom-animated-caption). - **You only want to switch presets on captions that are already on your timeline.** Click the caption track in the editor and use the **Caption style** menu directly; see [Caption Styles](https://pr-824-preview.chatcut.dev/docs/caption-styles#how-do-i-apply-a-style). - **You need to translate captions into another language first.** Add Captions styles captions in the video’s spoken language. Use the subtitle translation workflow in [Transcript and Captions](https://pr-824-preview.chatcut.dev/docs/transcript-and-captions) before or after styling. ## What do I need before I start? - **One spoken video.** Upload or select 1 video in the Skill form, or reference a video already in your project. - **A chosen caption style.** Pick one of the 49 built-in presets across the **Featured**, **Regular** (ordinary fixed styles), and **Dynamic** (animated word-timed styles) tabs. Every built-in style works in ChatCut Web and ChatCut Desktop except **Ember Scribble** (Web only), and animated **Dynamic** styles require ChatCut Desktop 0.8.7 or later. - **Optional confirmed spellings and layout notes.** Prepare any verified names, brands, or numbers (up to 500 characters) and placement or formatting notes (up to 2,500 characters). You can also attach up to 1 optional reference image when offered. - **Credits.** The Add captions Skill does not require a paid Pro plan. Running it with the built-in Agent uses ChatCut credits for the Agent request, while speech transcription, applying a built-in preset, and manual caption edits do not add a separate media-generation charge. A connected external agent uses its host billing for reasoning. See [Credits](https://pr-824-preview.chatcut.dev/docs/credits-policy). ## How do I run it? 1. Open [Add Captions](https://pr-824-preview.chatcut.dev/skills/add-captions), or choose **Skills → Add captions** in the AI composer (in ChatCut Desktop, you can also type `/` to search for it). 2. Upload or select your video and browse the **Featured**, **Regular**, and **Dynamic** tabs to choose a preset card. 3. Optionally enter confirmed spellings for proper nouns or numbers and any placement notes, then select **Apply this style** (or send your message in the composer). 4. Wait for transcription and preset application. Because you already chose the exact style in the form, the Agent applies that preset ID directly without asking for a second style picker. 5. Inspect the named editable timeline in the Viewer. Play through key lines to check word timing, spelling, and placement over your footage. ### Fill in the form This is the same form you see on the [Add Captions Skill page](https://pr-824-preview.chatcut.dev/skills/add-captions) and in ChatCut. Try it here: it opens ChatCut with your choices filled in. Try the form Try the Add captions form on its [Skill page](https://pr-824-preview.chatcut.dev/skills/add-captions). [](https://pr-824-preview.chatcut.dev/skills/add-captions) | Field | Options | What it changes | | --- | --- | --- | | **Your video** (required) | Upload or select 1 spoken video | The source video to transcribe and caption | | **Caption style** (required) | Tabs **Featured**, **Regular**, and **Dynamic** (49 built-in styles) | The exact canonical preset ID applied to your caption track | | **Confirmed spelling** (optional) | Free text, up to 500 characters | Verified proper nouns, brand names, or numbers to apply when correcting speech-recognition words | | **Additional direction** (optional) | Free text, up to 2,500 characters | Extra instructions for placement or layout, such as keeping captions clear of a lower third | | **Reference image** (optional) | Up to 1 image, when offered | Visual context for placement or layout | ### How the form preview works vs. the editor menu In the **Add captions** Skill form, preset cards animate using **sample words**. Even when you scrub to a frame of your uploaded video as a viewing aid, the form preview still displays sample text rather than your video’s transcript, and that preview frame timestamp is never used to trim your footage. Once the Skill places captions on your timeline—or when you open **Caption style** / **Caption presets** from the editor toolbar on an existing caption item—ChatCut previews and renders styles using your video’s actual transcribed words. ## Prompts you can copy Prompt ```text Use Add captions on @talking-head-clip with the Kinetic Poster style. Confirmed spelling: ChatCut, Kiyomi Tanaka, Q3 2026. Keep the captions in the lower third clear of my chin and do not export yet. ``` Prompt ```text Run Add captions on @product-demo and apply the Netflix preset. Keep the original 16:9 aspect ratio and spoken audio intact. ``` Prompt ```text Apply the Prime animated caption style to @podcast-intro using Add captions. Correct the speaker name to Dr. Elena Vance whenever it appears in the transcript. ``` ## What does ChatCut do for me? 1. **Reads the source video and word-timed transcript.** The Agent transcribes the spoken audio (or reads the existing transcript) while preserving your original words, audio track, and source duration. 2. **Applies confirmed spelling corrections.** It uses your submitted spelling list only for confirmed speech-recognition fixes on names, terms, or numbers, without rewriting what was spoken. 3. **Applies the exact preset ID.** It enables captions on a named editable timeline and applies the exact built-in preset ID you selected. It does not substitute a similar style or reinterpret your selection as a generic clean, emphasis, or word-by-word look. If a preset is unavailable on your current client (such as **Ember Scribble** on Desktop), it reports that concrete mismatch. 4. **Verifies composed frames.** It inspects representative frames over the actual source video to verify the preset appearance, readable placement, source aspect ratio, word timing, and original speech. ## What do I get? By default, you get a **named editable timeline** containing your full source video in its original aspect ratio with styled, word-timed captions enabled. From there, you can edit individual words or timing in the **Transcript** panel, tweak fonts, colors, or placement in the Inspector, or switch presets in the **Caption style** menu. ## What it does not do - **Does not generate an automatic uncertainty checklist.** The Skill does not output a confidence-score list of uncertain proper nouns or numbers. Listen to the original audio in the editor to verify unfamiliar names or figures, and supply confirmed terms in the form or edit them in the **Transcript** panel. - **Does not translate captions into another language.** It styles captions in the spoken language of the recording. To translate subtitles, use [Transcript and Captions](https://pr-824-preview.chatcut.dev/docs/transcript-and-captions). - **Does not trim your video to the preview frame or replace speech with sample text.** The form’s preview text and frame scrubber are viewing aids only. - **Does not automatically export SRT, TXT, or MP4 files.** An editable timeline is the default deliverable; ask the Agent to export or use **Export** in the editor when you want a file. See [Export Video, Audio, and Subtitles](https://pr-824-preview.chatcut.dev/docs/export-video-audio-subtitles). ## What related pages can I read? - [Add Captions Skill page](https://pr-824-preview.chatcut.dev/skills/add-captions) - [Caption Styles](https://pr-824-preview.chatcut.dev/docs/caption-styles) - [Transcript and Captions](https://pr-824-preview.chatcut.dev/docs/transcript-and-captions) - [Speech Cleanup](https://pr-824-preview.chatcut.dev/docs/speech-cleanup) - [Export Video, Audio, and Subtitles](https://pr-824-preview.chatcut.dev/docs/export-video-audio-subtitles) - [Credits](https://pr-824-preview.chatcut.dev/docs/credits-policy) - [All Skills](https://pr-824-preview.chatcut.dev/skills)