Skip to main content

How to translate a video to English

To translate a video to English, turn its speech into text, translate the text, then either show it as English subtitles or have an English voice read it over the picture. Subtitles are the cheapest and keep the speaker's real voice, a dub suits viewers who won't read, and lip-sync is only worth its price when a face is large on screen.

By Sarthak Daruka · Updated · 5 min read

In this guide
  1. Subtitles, dub or lip-sync: which do you need?
  2. How to add English subtitles
  3. How to dub a video into English
  4. What it costs to translate a video to English
  5. How to check the translation before you publish
  6. Whose voice and face can you use?
  7. If the video lives on YouTube

Subtitles, dub or lip-sync: which do you need?

The three do different jobs and the price gap between them is wide, so decide before you upload.

Three ways to translate a video to English
Best forThe catch
SubtitlesInterviews, talks, anything where the speaker's own voice mattersPeople have to read, and fast speech has to be shortened to fit
DubTraining, product demos, voiceover on slides or B-rollLips won't match if a face is on screen
Dub with lip-syncPresenters, ads, founders talking to cameraIt costs many times a plain dub

A quick test: watch the video with the sound off and notice how long a mouth is clearly on screen. If that's most of the video, lip-sync earns its price. If it's mostly slides, hands and product shots, a plain dub looks the same for a fraction of the cost. In Pleosa, lip-sync on Auto makes that call piece by piece: it re-syncs the lips only where a face is on screen, and the pieces without one cost less.

How to add English subtitles

Use Accessibility Studio when you want subtitle files and no new voice. It takes recordings of up to 2 hours.

  1. 1.

    Upload the video

    Open the studio, choose Captions and drop in the video. Only the speech is uploaded, compressed on your device. Leave Spoken language on Detect automatically unless it guesses wrong.

  2. 2.

    Make the captions

    Press Make captions. The price is on the button once the file is read.

  3. 3.

    Fix the original first

    Read the captions in the original language and correct names, product terms and numbers before you translate. A mistake here gets translated too, and it's harder to spot in English.

  4. 4.

    Translate

    Under Translated subtitles, set Translate into to English and press Translate. The timings stay the same, and you can edit the English lines like the original.

  5. 5.

    Download

    Take an SRT or WebVTT file for your player or platform. For a platform with no captions upload, make a copy with the captions burned into the video: it's free and made on your device.

SRT and WebVTT hold the same thing, timed lines of text. WebVTT is the format browsers use for text tracks on web video; SRT is the older one that most editing software takes. Keep a file even if you burn the subtitles in, because people can switch a file off and a burned-in line can't be corrected later.

How to dub a video into English

Use the AI Video Translator. It takes an MP4, MOV or WebM of up to 10 minutes, and your video's music and background sound stay under the new voice.

  1. 1.

    Upload a clean version

    Under Your video, choose or drop the file, and keep the tab open until the upload finishes. Use a version without burned-in captions: they'd stay in the original language, and lip-sync can smudge text near the mouth.

  2. 2.

    Choose English

    Pick English under Translate into. One run can take up to 10 languages, each with its own price.

  3. 3.

    Quick or Review script

    Quick transcribes, translates and voices in one step. Review script lets you read and edit every line before anything is voiced. Choose it for anything with names, numbers or claims in it.

  4. 4.

    Choose the voice

    Matched voices gives each speaker a natural stock voice close to their own. Their own voices matches each speaker's voice from this video. My voice uses a verified voice from your Voice Vault for the main speaker.

  5. 5.

    Set the lip-sync

    Auto re-syncs only where a face is on screen, Always does every piece, Off is voice only. Then pick the quality: Standard for most videos, Premium for sharper lips and teeth on medium shots, or Close-up when the face fills the frame.

  6. 6.

    Check the price and start

    Each language's price shows before anything is charged. Confirm whose face and voice it is, then start. Once it's running you can close the tab, and you'll be notified when it's ready.

  7. 7.

    Download

    Take the MP4, the SRT or VTT subtitles and the transcript, or make a copy with burned-in subtitles. Subtitle files are included in the price of every language.

What it costs to translate a video to English

Typical credits for one language
A clip up to a minuteEach minute of a 10-minute video
Subtitlesabout 2about 1
Dub, voice only67
Lip-sync, Standard4145
Lip-sync, Premium93103
Lip-sync, Close-up261268
From the functions that charge you, with every piece lip-synced, so Auto costs less when faces come and go. Subtitles are priced by the length of the speech and the amount of text. Your video's exact price shows before anything is charged.

Another language from the same upload costs less, about 3 credits a minute for voice only, because the transcript and the voice separation are paid for once. If a piece's lips can't be re-synced, it gets the new voice over the original picture and its lip-sync share is refunded automatically. See plans and credit prices for what a credit costs.

How to check the translation before you publish

A translation can read well and still be wrong. Check it with the picture running, not as a document, because a line that's right on paper can land on the wrong shot.

  • Names and numbers first. People, products, prices, dates and units are where errors cost the most. In the translator, the glossary keeps names exact, and your Brand Vault's brand and product names are in it while that switch is on.
  • Meaning, not grammar. Jokes and idioms rarely cross over intact, and many languages mark formal and informal speech where English doesn't. Check the tone survived.
  • Subtitle speed. Netflix's English style guide allows up to 20 characters a second for adult programmes, 42 characters a line and two lines at most. The BBC recommends 160 to 180 words a minute. Over that, cut words instead of stretching the time.
  • Dub timing. In Review script, lines too long for their moment are flagged. Shorten them: a rushed line is the first thing that gives a dub away.
  • A fluent reader. For legal, medical, financial or safety content, have someone fluent in both languages sign it off.

Whose voice and face can you use?

The tool asks you to confirm whose face and voice it is before anything is charged. Their own voices is a one-time match made from your video, and the sample is removed after an hour.

If the video lives on YouTube

You have two routes. Upload the English version as a video of its own, or keep one video and add English as a second audio track. YouTube's multi-language audio is open to creators with access to its Advanced features and takes an audio-only file about as long as the video, so you'd export the sound from your translated MP4 first.

Try it on one short clip

Upload a minute of video and you'll see the exact price before anything is charged. Subtitle files come with every language.

Open the AI Video Translator

Sources

  1. Netflix, English (USA) Timed Text Style Guide. Read 2 November 2026.
  2. BBC, Subtitle Guidelines. Read 2 November 2026.
  3. W3C, WebVTT: The Web Video Text Tracks Format. Read 2 November 2026.
  4. YouTube Help, Add Multi-language features to your videos. Read 2 November 2026.

Keep reading