> For the complete documentation index, see [llms.txt](https://learn.firecut.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://learn.firecut.ai/features/add-captions.md).

# Add captions

## ❓ What is it?

FireCut lets you add captions to your video.

While English is fully supported, **FireCut also lets you transcribe audio in 50+ languages**; however, please do double-check the results because this is an experimental feature.

## 🤔 How does it work?

The "Add captions" feature:

* Transcribes your audio and gets accurate timings for every word, **all done automatically**. Longer sequences are done in parts of about 5 minutes each, so you'll see **Transcribing... (1 of 3)** and so on
* Lets you modify the captions in 2 different modes:
  * **Styled captions** (if you want captions that are styled and animated within FireCut), or
  * **Text-only captions** (if you just want a captions track that you can style yourself in Premiere Pro)
* You can also export the captions as SRT, VTT or TXT files

{% embed url="<https://www.youtube.com/watch?v=NkCEjmI3Mbk>" %}

### 1️⃣ Styled captions mode

* **When to use:** To easily get animated captions that match popular creators' styles
* **Output:** Caption images (.PNG files) inserted in a new video track in your sequence, as a single nested sequence by default (double-click it to see the individual caption clips). Even though these are images, [**they can still be edited by FireCut**](#editing-styled-captions)

### 2️⃣ Text-only mode

* **When to use:** To get a text-only captions track that you can edit in Premiere Pro
* **Output:** New captions track (requires Premiere Pro 2022 or later) that contains your transcript. To change its text, use Premiere Pro's **Text** panel

## ⚒️ Styling your captions

After you click **Transcribe**, FireCut opens the captions editor in **STYLED** mode. The preview at the top shows your captions on your video. Click the preview to move the captions, or use **Reset X** and **Reset Y** to centre them again. On vertical (9:16) sequences, **Safe zone** shows where TikTok, Instagram, Facebook and YouTube Shorts place their buttons.

**Edit transcript** is where you fix the words:

* Click a word to edit it, and click the **/** between words to add or remove a line break
* Use **Find / Replace**, **Undo**, **Redo** and **Copy** as you would in a text editor. **Reset edits** takes the transcript back to how it was transcribed
* Turn on **Emphasize words** and click words to give them an emphasis style, or click **Detect words** to have FireCut pick them
* Click **Detect emojis** to have FireCut suggest emojis for your captions
* **Timestamps** and **Text size** only change how the transcript looks in the panel, not your captions

Below the transcript, these sections change how your captions look:

* **Presets:** Click a preset to load its whole style. See [Saving your brand style](#saving-your-brand-style)
* **Font:** Font name, size, weight, alignment, colour (solid or gradient), italic, and letter, word and line spacing
* **Emphasize words:** The style for your emphasised words. Click **+ Add emphasis style** if you want more than one, or **Detect words with AI** to pick the words
* **Outline**, **Box** and **Shadow:** An outline around the text, a box behind it and a drop shadow
* **Emojis:** Size, position (above or below the captions), library and animation
* **Wiggle** and **Tilt:** Shake your captions, or tilt each one at a random angle
* **Remove profanity:** Replace swear words with stars or symbols, and mute or bleep them in the audio
* **Animation:** Pop, reveal, grow, bold, highlight, glow, outline, underline or box each word as it's spoken, plus blur and fade transitions
* **Multiple lines:** Show more than one line of captions at a time
* **Adjust line breaks:** Leave this off and FireCut breaks lines to fit the font size and sequence width. Turn it on to set **Words per line** (1 to 20, or no limit) and **Characters per line** (1 to 100, or no limit). **Force line break on pause** (after 1 second by default), **Force line break on full stop** and **Keep related words together** are on by default
* **Adjust timing:** **Extend to next word (within a caption)**, **Extend to next caption** and **Align captions to cuts**
* **Adjust formatting:** **Remove punctuation** and **Convert to case** (**lowercase**, **(no change)** or **UPPERCASE**)
* **Merge repeated words:** Turns "More more more" into "More X3"

**Emojis**, **Wiggle**, **Tilt**, **Apply fade** and **Add "pop" animation to box** don't work with the **Turbo** rendering algorithm.

### Styling more than one speaker

If you transcribed with **Multi-speaker mode**, pick a speaker under **Styling** first. Presets and every control below it apply only to that speaker, so each speaker keeps their own style. You can also:

* Turn on **Speaker names** to show each speaker's name above their captions, with its own colour, size and gap
* Use **Position per speaker** above the preview to place each speaker's captions separately
* Click a speaker under **Show lines from** to see only their lines in the transcript
* Click the coloured speaker chip at the start of a line to give that line to a different speaker

**Emojis** aren't available for multi-speaker captions, because each speaker is styled separately.

## ⚒️ Adding styled captions to your sequence

When your captions look right, click **Add styled captions to sequence**. The settings just above that button decide how they're added:

1. **Caption type:** **Nested sequence** (the default) adds your captions as a single clip, which is faster. Double-click it to open the sequence with the individual caption clips. **Individual clips (legacy)** adds each caption as its own clip, which is easier if you want to fine-tune single lines or words in the timeline. You can edit both with FireCut later
2. **Rendering algorithm:** Only shown for **Individual clips (legacy)**. **Rapid (Default)** suits most sequences. **Turbo (much faster for long sequences)** is experimental: it makes a copy of your sequence with a new track for your captions, and some effects don't work with it. **Original (Legacy)** is the older, slower method
3. **Use multicore rendering:** Uses more of your computer's CPU cores to make the caption images faster. It's experimental and off by default. If you turn it on and your captions look different from the preview (e.g. the wrong font), turn it off again
4. **Target tracks:** Where the captions go. The default (and safest) choice is new track(s). If you pick tracks that already have clips on them, those clips may be deleted and replaced by the captions. FireCut warns you before this happens
5. **Save location:** Where the caption images and data are saved: **Save to project directory** or **Choose location** (the default). Keep this folder where it is if you want to edit the captions later

If you'd rather have a subtitle file, click the **...** button next to **Add styled captions to sequence** and choose **Export as SRT**, **Export as VTT** or **Export as TXT**.

## ⚒️ Text-only captions

Click **TEXT-ONLY** at the top of the captions editor to switch modes. There are 3 settings:

1. **Adjust line breaks:** Leave this off to keep the transcript's own breaks, which are usually one phrase per caption. Turn it on to set **Max words per line** and **Max characters per line**, and to choose whether to **Force line break on full stop**
2. **Adjust timing:** Turn on **Extend to next caption** to keep each caption on screen until the next one starts
3. **Adjust formatting:** **Remove punctuation**, and **Convert to case**: **lowercase**, **(no change)** or **UPPERCASE**

Then click **Add captions track to sequence**, or **Export as SRT...**, **Export as VTT...** or **Export as TXT...**. Click a caption in the preview list to jump to that point in your sequence. To change the text after you've added the track, use Premiere Pro's **Text** panel.

## 💾 Saving your brand style

Once you have changed the various settings (e.g. font, text position, animation) to match your style, just **click one of the \[+] buttons** to save this style as a Preset. Next time, you can just click the name of the Preset to load the style again.

<div align="left"><figure><img src="https://4122560492-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FWDn6n1C7MKZBHdDcmm6f%2Fuploads%2F5hrWJyopfpO7znVxYxx3%2Fimage.png?alt=media&amp;token=eaea24f4-a27d-4230-bf8d-56dd3113d4eb" alt="" width="375"><figcaption></figcaption></figure></div>

## ⚒️ Sharing your style/preset with others

Once you have saved your style as a preset using the instructions above, you can also share it with other FireCut users. By default, these are shared with the FireCut community (i.e. they are available to all FireCut users), but if you're part of a team you can choose to share it only with your team.

To share a preset:

* Click on "**Edit**" to see your list of custom presets\
  ![](https://4122560492-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FWDn6n1C7MKZBHdDcmm6f%2Fuploads%2FJBXglicPAXOVUIMKQYqE%2Fimage.png?alt=media\&token=32b35de5-f659-430d-a7f6-3d0324e40fd6)
* Click on the share icon next to the preset you want to share![](https://4122560492-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FWDn6n1C7MKZBHdDcmm6f%2Fuploads%2FO7BDV1d8a3r7tkcCsMCd%2Fimage.png?alt=media\&token=e200bce6-99c2-4ede-98b0-5a58be1e6af4)
* Enter the **Preset name**, and a **Creator name** (your name as you'd like other users to see it). If you're part of a team, choose **Public** or **Team** under **Visibility**. Then click "**Publish**"\
  ![](https://4122560492-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FWDn6n1C7MKZBHdDcmm6f%2Fuploads%2FHbSfzVq6dPplk7W6WYUE%2Fimage.png?alt=media\&token=1e8117e0-5053-48e3-80c1-cf14f832bd68)

Your preset is visible to others straight away. Once it's shared, you can't unshare it from the panel, and deleting it from your list only removes your own copy. If you need a shared preset taken down, email us at <support@firecut.ai>.

To use other people's presets, click **Browse community presets** in the **Presets** section. Pick the **Public** or **Team** tab, select a preset and click **Save to your library**. It's added to your own presets.

## ⚒️ Editing styled captions

Styled captions are inserted as .PNG images in your sequence. This is done by design to give you a great deal of flexibility in how the captions are styled and animated.

While these are .PNG images, **it is still possible to edit them using FireCut.**

1. Select the caption clips in the timeline. This works for both nested caption clips and for individual caption clips
2. Go to **FireCut** > **Add captions** and click **Edit selected captions**
3. Make your changes in the **TRANSCRIPT** and **STYLE** tabs
4. Click **Update captions**

<figure><img src="https://4122560492-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FWDn6n1C7MKZBHdDcmm6f%2Fuploads%2Fgit-blob-2b6e0d3fcb5cd4b2274252b0b24f4837a70e7a48%2Fadd-captions-start.png?alt=media" alt="" width="375"><figcaption></figcaption></figure>

FireCut loads the part of your transcript that matches the captions you selected. If you want to reload the full transcript instead, click the **Load full transcript** button.

<figure><img src="https://4122560492-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FWDn6n1C7MKZBHdDcmm6f%2Fuploads%2Fgit-blob-8c2ebcd5ef9d6b2538636b05c60638a96eda7f21%2Fadd-captions-edit-mode.png?alt=media" alt="" width="375"><figcaption></figcaption></figure>

A few things to know when you edit:

* If you only change words, FireCut swaps just those captions in place. If you change the style or line breaks, it adds the loaded captions again in the new style
* Your caption tracks are muted while you edit, so they don't show twice in the preview. FireCut turns them back on when you leave the editor
* Select either nested captions or individual caption clips, not both. Nested captions also need to come from the same captions sequence
* If the selected captions use different styles, the **STYLE** tab is locked. Select only the captions that share a style and edit those
* **Edit selected captions** needs the captions data FireCut saved next to the caption images. If that folder was moved or deleted, the button stays greyed out
* Translating and the output settings aren't available while you edit captions that are already in your sequence

## ⚒️ Translating your captions

You translate your captions after they're transcribed. Set **Audio language** to the language your audio is in, click **Transcribe**, then:

1. Under **Edit transcript**, click **Choose language...** next to **Translate to another language**
2. Pick a language under **Translate to**, or click one of the **Popular** or **Last used** languages
3. Turn on **Show below original** if you want to keep the original captions and add the translation as a second line underneath
4. Click **Translate**

{% hint style="info" %}
Translating locks your line breaks, so each translated caption stays paired with the words it came from. The line-break and animation settings are unavailable until you click **Reset** (next to **Choose language...**), which brings back the original captions.
{% endhint %}

* When both languages are shown, you can change the size of the second line and the gap above it under **Translated line**
* Multi-speaker captions are always translated in place, so **Show below original** isn't offered for them
* You can't translate while you're editing captions that are already in your sequence. Load the transcript from [History](#history) instead, translate it and add the captions again

## ⚒️ History

Every transcript you make is saved under **History**, below the **Transcribe** button. Each row shows when it was made, what was done (e.g. Transcribed), the sequence name and the first few words.

* Click a row to open that transcript in the captions editor again, without transcribing your audio again
* Use the search box to find a transcript by sequence name or by any words in it
* **Clear** deletes your whole captions history. FireCut asks you to confirm first

Transcripts made with **Multi-speaker mode** only show in History if Multi-speaker mode is available on your plan.

## ⚙️ Settings

There are 6 main settings for add captions:

1. **Scope:** You can decide whether to run the operation on your **Full sequence** or only on a certain portion that you specify with **In / Out points** (you can place these in your sequence using the `I` and `O` hotkeys)
2. **Audio language:** The language of your audio. We are constantly adding support for new languages. However, please note this is experimental and may create unpredictable results when compared with English. This is the case for every AI tool even if they don't mention it, because [training sets for AI tend to be in English](https://www.wired.com/story/chatgpt-non-english-languages-ai-revolution/). On occasion, this might cause the operation to fail. Please do let us know (<support@firecut.ai>) if this happens, so we can improve it for you! You can set a default language in FireCut's **Settings** menu so you don't have to change it every time. Choose **English (British)** if you want British spellings in your captions
3. **Guidance prompt:** An optional text box that guides how your audio is transcribed. You can include:

   1. **Brand names / Proper nouns** that may not be spelled correctly by default (e.g. type "FireCut" in order to correct "fire cut" to "FireCut" in the transcript). This is what the guidance prompt does best
   2. **Capitalisation preference**, for example, if the transcript is coming out in lower case ("the park is green") and you want sentence case ("The park is green."), write a short sentence here in your preferred capitalisation and FireCut will try to follow it
   3. **Punctuation preference**, for example, if the transcript tends to miss punctuation, write a short sentence here with the level of punctuation you want and FireCut will try to follow it

   To reuse a prompt, click **+** next to **Presets:** to save it, then click its name next time
4. **Use voice activity detection:** If you have long silences in your audio, they can cause problems in the transcript, such as repetitions of the same word or words that were never said. Voice activity detection (VAD) helps to avoid this by skipping the silent parts. It's on by default
5. **Multi-speaker mode:** Labels each speaker in the transcript so you can style their captions separately (see [Styling more than one speaker](#styling-more-than-one-speaker)). This is experimental. Choose where FireCut finds the speakers:
   * **Mixed audio:** FireCut listens for the different voices in your mixed audio. Leave **Number of speakers** on **Auto**, or set it from 2 to 5
   * **One track per speaker:** For when each person is on their own audio track, like a podcast with one mic per guest. Pick each speaker's **Audio track** and type a **Speaker name**, and click **+ Add speaker** for more. You need 2 to 5 speakers, each on a different track, and each track needs clips between your In and Out points
6. **Show video in preview window:** Plays your video in the captions preview instead of a still frame. FireCut exports up to the first 90 seconds of your sequence for this, so transcribing takes a little longer. It's off by default

## 💡 Tips for best outcomes

* **Make sure your audio does not have long silences**, and any music / sound effects are muted
* **Remove any repetitions first**, as FireCut might not transcribe multiple instances of the same text
* **Don't work in the timeline** while FireCut is working in the background
* **Finish your edit before you add captions**. FireCut's transcription follows your **sequence**'s timecode, not individual clips, so if you move clips afterwards, the captions won't move with them
* **Load a transcript from History** instead of transcribing the same audio again

## 🙋‍♂️ Common questions, issues, watch-outs

* **"The transcribed text was wrong"** -->
  * **All languages:** Please make sure the audio is clear, there are no long silences, and no repetitions. If this doesn't fix the issue, please share your audio with us at <support@firecut.ai>
  * **Non-English:** Transcription for non-English audio is currently experimental and expected to be 70 to 80% correct in most cases. If you're experiencing a much worse outcome, please get in touch and share your audio with us at <support@firecut.ai>
* **"The transcript has a style I don't like" (e.g. too many capital letters, everything in lowercase, missing punctuation, etc.)** -->
  * Use the "guidance prompt" setting (see [#settings](#settings "mention") above) to guide the transcript's style next time you transcribe
  * For a transcript you already have, use **Find / Replace** in **Edit transcript**, or **Adjust formatting** to remove punctuation or change the case
* **"I can't find the setting to translate my captions"** --> Translating happens after you transcribe, under **Edit transcript**. See [Translating your captions](#translating-your-captions)
* **"The same person has different speaker labels in different parts of my video"** --> On longer sequences, **Mixed audio** can give the same person a different label in different parts. If each person is on their own audio track, transcribe again with **One track per speaker**. Otherwise, click the speaker chip at the start of a line to give it to the right speaker
* **"Edit selected captions is greyed out"** --> Select FireCut caption clips (or a captions nested sequence) in your timeline first. If they're selected and the button is still grey, FireCut can't find the captions data it saved next to the caption images, usually because that folder was moved or deleted
* **"My caption tracks were muted"** --> FireCut mutes them while you edit captions that are already in your sequence, so they don't show twice in the preview. They're turned back on when you leave the editor
