# Captions and transcription Turn audio or video into text with Transcribe, and add timed captions to a video or sound clip in the AI Editor. Source: https://infinall.ai/docs/audio/captions-and-transcription Last checked: 2026-10-26 ## What it is There are two tools. **Transcribe** turns an audio or video file into text and keeps the timing of each line. **Captions** in the AI Editor listen to a clip on your timeline and add timed subtitles you can style and edit. ## Who can use it Everyone, on Free and Pro. Neither tool is locked to a plan in the code. The AI Editor opens on a laptop or desktop. On a phone it shows "The AI Editor works on a computer." ## Cost - **Transcribe** uses credits. It is charged by the length of the audio, and for a video it uses the audio track. You see what it cost on the result. - **Captions** in the AI Editor run on your computer and your audio is not uploaded. Making subtitles from a clip is free in credits. The **Let AI pick the look** switch is a paid step, and its price is shown on the switch itself. Turn it off to stay free. - Importing an SRT or VTT file is free. See [Price before you confirm](/docs/credits/price-before-you-confirm). ## Where to find it - **Transcribe:** on **Home**, choose the **Audio** tab, open the mode chip and pick **Transcribe** in the Words group ("Audio or video to text"). You can also use it from the Audio composer on the Studio canvas. - **Captions:** open **AI Editor** in the sidebar, click a video or sound clip on the timeline, open the **Effects** view of the inspector (it has the **AI** controls), and open the section **Local Auto-Captions**. ## Buttons and controls ![The Transcribe mode on Home with the upload area for audio or video files.](/docs/screens/audio/captions-and-transcription.webp) Transcribe, in the Audio composer: | # | Name in the app | What it does | Plan | | --- | --- | --- | --- | | 1 | Transcribe | The mode chip, after you pick Transcribe. | Both | | 2 | Upload audio or video to transcribe | The upload area. Click it or drop a file. It lists the file types and the size limit. | Both | | 3 | Start campaign | The round arrow that sends the file. On the Studio canvas it is named **Generate audio**. | Both | | | Remove file | The X on a picked file, shown once you pick one. Takes the file out so you can pick another. | Both | | | Copy | On the **Transcript** card on the canvas. Copies the text. | Both | Captions, in the AI Editor, inside **Local Auto-Captions**. These controls are not numbered because the screenshot above shows Transcribe: | # | Name in the app | What it does | Plan | | --- | --- | --- | --- | | | Local Auto-Captions | The section title. It is closed at first, so open it. | Both | | | Clip | Step one. Says "Click a video or sound clip on the timeline." until a clip is selected, then "Clip selected. Ready to listen." | Both | | | Spoken language: English | Opens the list of spoken languages. The button name changes with the language you pick. | Both | | | Let AI pick the look | A switch. When on, it places and styles the subtitles to suit your video. It is on at first. | Both | | | Punch, Blaze, Wave, Clean, Pop, Bold | Six looks to pick from. Hover for what each is best for. | Both | | | Make subtitles | Listens to the clip and makes the subtitles. While it works it reads **Listening...** with a percent. | Both | | | Add subtitles to video | Appears after listening. Adds the subtitles to the timeline. The button shows the number, for example **Add 12 subtitles to video**. | Both | | | Import SRT / VTT as Text | Opens a file picker for a subtitle file. Each caption becomes an editable text clip. | Both | | | Editable Captions | A section below. Edit caption text, select captions, and split them into single lines. | Both | | | Maximum words per caption | A list inside Editable Captions, and used when making subtitles. Picks how many words fit on one caption. | Both | | | Make all single-line, Make selected single-line | Splits each caption into short timed clips, good for vertical video. | Both | | | Select all, Clear selection | Picks or clears all captions in the list. | Both | ## How to transcribe a file 1. On **Home**, choose the **Audio** tab. 2. Open the mode chip and choose **Transcribe**. 3. Click **Upload audio or video to transcribe** and pick a file, or drop one on it. A picked file shows its name, size and either "ready to transcribe" or, for video, "the audio track will be transcribed". 4. Send it with the round arrow. 5. A **Transcript** card appears on the canvas. Select the words, or press **Copy**. Each line keeps its start and end time, so the transcript can be used as captions. ## How to add captions to a video in the AI Editor 1. Open **AI Editor** and click a video or sound clip on the timeline. 2. In the inspector, open the **Effects** view, then open **Local Auto-Captions**. 3. Choose the spoken language with **Spoken language**. 4. Leave **Let AI pick the look** on, or turn it off and pick one of the looks yourself. 5. Choose **Make subtitles** and wait for the progress to finish. 6. Read the list ("Here is what we heard. Check it, then add it to your video.") and choose **Add subtitles to video**. 7. To tidy the text, open **Editable Captions**, edit any caption, and use **Make all single-line** if you want one short line at a time. ## How to import a subtitle file 1. Select a clip, open **Local Auto-Captions**. 2. Choose **Import SRT / VTT as Text**. 3. Pick a `.srt` or `.vtt` file. You see "Captions imported" when it works. Each cue becomes an editable text clip on the Captions track. ## Options explained **Spoken languages for captions.** English, French, German, Spanish, Italian, Portuguese, Hindi, Japanese, Korean, Chinese, Russian, Turkish, Polish and Vietnamese. **Looks.** Each look has a best use, shown when you hover: **Punch** for talking head and business, **Blaze** for gaming and entertainment, **Wave** for music and sing-along, **Clean** for professional and minimal, **Pop** for fun and energetic, **Bold** for viral and high impact. **Let AI pick the look.** With the switch on, an "AI chose" note tells you where the captions were placed and which look was used, with a short reason. You can still tap another look. The switch only works for video clips. **Maximum words per caption.** A list of 10 choices, starting at 2 words. It starts at 5 and you can change it any time. **Transcribe file types.** Audio: MP3, WAV, M4A, OGG and WEBM. Video: MP4, MOV and WEBM. ## Limits - Files for Transcribe can be up to 50 MB. - The AI Editor needs a laptop or desktop. ## When something goes wrong | Message in the app | What it means | What to do | | --- | --- | --- | | Upload an audio file first. | Sent in a file mode with nothing picked. | Pick a file, then send. | | That file type isn't supported. Upload audio (MP3, WAV, M4A, OGG) or video (MP4, MOV, WEBM). | The file is not a type Transcribe can read. | Convert it, or pick another file. | | `That file is MB. The limit is 50 MB.` | The file is over the size limit. | Cut or compress it first. | | That file is empty. | The file has no content. | Pick another file. | | This is text, not audio, so it cannot be exported as a sound file. | You tried to export a transcript as sound. | Use **Copy** instead. | | No speech was detected in the selected clip. | Captions could not hear speech. | Pick a clip with clear speech, or change **Spoken language**. | | No speech was detected inside the selected clip. | The speech is outside the part of the clip you kept. | Widen the clip on the timeline, then try again. | | The selected clip audio could not be decoded. | The audio in the clip could not be read. | Re-import the media and try again. | | Reconnect the source media before creating captions. | The source file is no longer connected to the project. | Re-add the media, then try again. | | Caption import failed | The subtitle file had no valid captions or could not be read. | Check that it is a valid `.srt` or `.vtt` file. | | Insufficient credits. Purchase more to continue. | Your balance is too low for Transcribe. | See [Your credit balance](/docs/credits/credit-balance). | ## Questions ### Does Transcribe work on a video? Yes. It reads the audio track of an MP4, MOV or WEBM file and gives you the text. ### Are the captions in the AI Editor private? Yes. The panel says "Runs on your computer. Your audio is not uploaded." ### Are the timings kept? Yes. Transcripts keep the start and end time of each line, so they can be used as captions. ### Can I use my own subtitle file? Yes. Choose **Import SRT / VTT as Text**. Each cue becomes an editable text clip. ### Is Transcribe free? No. It uses credits, charged by the length of the audio. Making captions in the AI Editor from a clip is free. Only the **Let AI pick the look** step has a credit price.