# Captions: text, accents, style, position

> How to fix caption text, highlight key words, pick a design, the spoken-word highlight and the phrase entrance, color by speaker and place captions.

Source: https://docs.slicer.media/en/guide/editor/captions/

Slicer makes captions from recognized speech: each word appears exactly when it is spoken. On the **Captions** track they are split into phrases. In the editor you can fix any phrase, highlight key words in the accent color, change the font and design, choose how the spoken word is highlighted and how a phrase appears, move the captions in the frame or hide them. With the **Auto** position, Slicer places each phrase where the frame is free.

**Demo (Web):**

1. Click a phrase on the captions track — the phrase tools appear at the bottom
2. Press Enter — the “Caption” sheet opens with the phrase text
3. Fix a word and click “Done” — the caption updates in the preview
4. Click “Accent” — the phrase's words are in the accent color
5. Click “Style” and pick the “Karaoke” design
6. Pick the “Rounded” font and the “Mint” accent — the preview changes at once

## Fix the text

### 1. Open the phrase

Any of these opens the **Caption** sheet with that phrase:

-   select a phrase on the captions track and press **Edit text** or `Enter`;
-   on the web — double-click or right-click the phrase; on iPhone and iPad — long-press it;
-   with nothing selected: **Text** → **Edit caption** — the phrase at the playhead opens.

### 2. Change the text

The sheet shows the phrase's time and a field with its text. Fix a typo, a name, a term or the punctuation.

### 3. Click “Done”

Or press `Enter`. The caption updates in the preview right away.

**Words keep the timing of the spoken ones — timing and highlights stay. Empty text hides this caption.** How the text is laid onto the words:

-   the same number of words as before — each replaces its own;
-   fewer words — they are spread evenly across the phrase's time;
-   more words — several new words show together in place of one spoken word.

Caption text changes only the on-screen text — the voice in the reel stays the same. **Restore the original text** in the sheet brings back the recognized text of the phrase.

**Demo (iPhone):**

1. Tap a phrase on the captions track
2. Tap “Edit text”
3. Fix a word and tap “Done”

## Accent

**Accent** on a selected phrase turns the highlight of all its words on or off. Highlighted words take the style's accent color:

-   in the “Dynamic”, “Classic”, “Pop” and “One word” designs, highlighted words are colored;
-   in “Karaoke”, the color marks the word being spoken right now;
-   in “Box”, the accent color is the box under the spoken word;
-   “Minimal” has no highlights.

The AI already highlighted key words during the montage — you can change them. To highlight specific words, the [AI assistant](https://docs.slicer.media/en/guide/ai/assistant/) is handy: “Highlight the numbers and names as accents”. You can also choose how the spoken word is highlighted yourself — with the **Word highlight** setting (Caption options).

## Caption style

The **Caption style** sheet opens with **Style** on a selected phrase, **Captions** in the main set or **Text** → **Caption style**. Everything changes in the preview at once.

| Setting | Options |
| --- | --- |
| **My templates** | Saved styles: click one to apply it. Type a name and press ✓ to save the current style as a template |
| **Font** | Punch, Poster, Block, Wide, Editorial, Rounded, Script, Mono |
| **Accent** | Amber, Azure, Lime, Crimson, Fuchsia, Mint, Violet, White |
| **Captions** | A switch: show captions in the reel or not |
| Design | Seven options — from “Dynamic” to “One word” (all designs) |
| Size | Small, Medium, Large |
| Position | Auto — each phrase goes where the frame has the most free space (more), Top, Center, Bottom |
| **Word highlight**, **Entrance**, **Words on screen**, **Letter case** | How the spoken word is highlighted, how a phrase appears, how many words are on screen and in which letters (Caption options) |
| **Colour by speaker** | A switch: each speaker's words in their own color. Shown when the sources' speech has two or more speakers ([more](https://docs.slicer.media/en/guide/editor/captions/#speaker-colors)) |

A template stores the font, the accent and all the caption settings — including the word highlight, entrance, words on screen and letter case. It also appears in the new project wizard on the “Style” step. The same settings are in the project's **Settings** section — a change in one place shows up in the other right away.

### Designs

A design is the ready-made character of the captions: how many words are on screen, how the spoken word is highlighted and how a phrase appears. In the wizard its row is called **Style**. Click a design — the preview shows it on your reel right away.

| Design | What it looks like |
| --- | --- |
| **Dynamic** | Short 1–3 word phrases, key words pop in the accent color |
| **Karaoke** | The full phrase on screen, the spoken word is highlighted in the accent color in sync with speech |
| **Classic** | Calm 1–2 line captions without animation, the accent only on key words |
| **Minimal** | Small neat text with a soft shadow, no highlights |
| **Pop** | 2–3 words on screen: the spoken word pops out in the accent color while the others make room |
| **Box** | A line of up to six words: the spoken word sits on an accent box that slides from word to word |
| **One word** | One big word at a time, in sync with speech — maximum energy |

A ready-made combination with “Pop” is the **Pop** template on the wizard's “Style” step: the Punch font and the “Lime” accent.

## Caption options

Four options fine-tune how captions follow the speech — on top of any design. They are in the **Caption style** sheet, in the project's **Settings** section and on the **Captions** step of the new project wizard. Each one defaults to **As in the style** (letter case — to **As in the font**): the way the chosen design or font intends.

-   ****Word highlight**** — how the word being spoken is highlighted: **Color** — in the accent color; **Pop** — the word grows and lights up while the others smoothly make room; **Box** — the word sits on an accent box; **Underline** — an accent bar under the word; **None**.
-   ****Entrance**** — how a phrase appears on screen: **None** — at once; **Fade** — it fades in and out; **Pop** — with a springy pop; **Slide up** — it slides in from below; **Word by word** — words appear as they are spoken.
-   ****Words on screen**** — how many words to show at once: **One**, **2–3 words**, **Line** — up to six words in one line, **Phrase** — the whole phrase, up to two lines.
-   ****Letter case**** — **As in the font**: capitals if the font is designed for them; **CAPITALS** — all capital letters; **As spoken** — regular case, as in the transcript.

How to use them:

-   **Design first, details second.** Picking a design resets **Word highlight**, **Entrance** and **Words on screen** to **As in the style** — so the new design shows up the way it is meant to. Letter case, size, position and colour by speaker stay.
-   **Mix them freely.** For example, “Karaoke” with **Pop** as the word highlight: the full phrase on screen, and the spoken word grows. Or “Classic” with **Word by word**: calm lines whose words appear as they are spoken.
-   **Key words stay colored** in the designs where they are colored (Accent), and the spoken word is highlighted on top of the speaker's color too.
-   **Letter case changes only the captions** — the hook and plaques keep the case the font is designed for.

## Color by speaker

In an interview, a dialogue or a podcast, viewers follow the conversation more easily when different people's lines differ in color. **Colour by speaker** paints each speaker's words in their own color — in the preview and in the finished video.

**Demo (Web):**

1. With nothing selected, press “Captions” — the “Caption style” sheet with the “Colour by speaker” switch
2. Turn on “Colour by speaker” — in the preview the second speaker's words are in another color
3. Seek to a line of the first speaker — their words keep the usual color, key words are in the accent

### 1. Open the caption style

**Style** on a selected phrase, **Captions** in the main set, or the project's **Settings** section.

### 2. Turn on “Colour by speaker”

The switch is in the same sheet, among the caption settings. The preview shows the new colors right away.

How the words are colored:

-   the first speaker keeps the usual caption color;
-   the others get colors from the style palette in order — except the accent color and white, so their lines don't mix with highlighted words;
-   the accent and the spoken-word highlight (for example, the one in “Karaoke”) go on top of the speaker's color.

The switch shows only if Slicer told at least two speakers apart when it recognized the speech in the sources. It is off by default.

## Position in the frame

Where the captions sit is set by the position options in the **Caption style** sheet — **Auto**, **Top**, **Center** and **Bottom** — and by the **Position** row in the new project wizard. The last three keep every phrase of the reel in one place. Horizontally, captions always stay centered.

### Smart Auto

With **Auto**, Slicer places each phrase on its own — where the frame has the most free space at that moment:

-   **off the face.** Slicer checks where the face is during this phrase — taking the crop, zoom, face following and camera motion into account;
-   **off the reel's elements.** A phrase steers clear of the hook, plaques, titles and picture-in-picture windows while they are on screen;
-   **no needless jumps.** When the frame is free, captions stay in their usual place in the lower part of the frame. If a phrase had to move, the next phrase in a row goes to the same place, and the captions return to the usual place on a cut — together with the change of picture;
-   **inside the safe area.** A phrase doesn't go to the very edges of the frame, where social apps cover the video with their buttons and captions;
-   **together with emoji and cards.** Emoji and media cards go next to the captions — wherever the captions are at that moment.

The preview and the finished video place phrases the same way: where a phrase sits in the editor is where it will be in the file.

Slicer knows where the faces are after an AI montage or after [Follow the face](https://docs.slicer.media/en/guide/editor/frame/#follow-face) on a clip. Until faces are found — for example, in a [project without the AI director](https://docs.slicer.media/en/guide/projects/manual/) — Auto steers clear only of the reel's elements: the hook, plaques, titles and picture-in-picture windows.

### Move manually

**Position** on a selected phrase opens the **Elements** mode with the hint **While paused, drag captions and elements right on the preview.** Pause the preview and drag the captions up or down — horizontally they always stay centered. Captions moved manually stay at that height for the whole reel: Auto no longer moves them.

After a manual move, the style sheet shows **Captions moved manually** and a **Reset position** button — the captions go back to the chosen position, and with **Auto** each phrase finds its own place again.

## Cut out and hide

> **Important.** **Cut out** (and `Delete` on a selected phrase) removes the phrase **from the reel entirely** — with the voice and the picture, not just the text. To hide only the text of one phrase, clear it in the **Caption** sheet and click **Done**. **Restore cut words** on the clip or undo brings the cut words back.

**Hide captions** turns captions off for the whole reel, **Show captions** turns them back on. It's the same switch as **Captions** in the style sheet.

## Good to know

-   [Freeze frames](https://docs.slicer.media/en/guide/editor/freeze-replace/) and reversed clips have no captions.
-   A copy of a clip with speech has no captions — they stay with the original ([Clips](https://docs.slicer.media/en/guide/editor/fragments/)).
-   To translate captions into another language, see [Caption translation](https://docs.slicer.media/en/guide/editor/translate/).
-   The director in the Chat changes the caption design and options too — for example: “Make the spoken word pop” ([Chat with the director](https://docs.slicer.media/en/guide/ai/chat/#style)).
-   Caption edits are free. They appear in the finished video after **Export**.

## What's next

-   [Caption translation](https://docs.slicer.media/en/guide/editor/translate/)
-   [Elements: hook, plaques, cards, emoji, zooms](https://docs.slicer.media/en/guide/editor/elements/)
-   [Frame: follow the face](https://docs.slicer.media/en/guide/editor/frame/#follow-face)
-   [AI assistant](https://docs.slicer.media/en/guide/ai/assistant/)