Skip to main content

Translate a video’s audible speech into a selected language and sync its lips with HeyGen Video Translate. Supply video_url, select from 178 target options and choose precision or speed. Permission is required from anyone made to appear speaking the translation.

From $0.048/s

Workbench

Input

Video translation
affects price1/1

Drop a video here, or browse · 1 file · Max 90 MB · MP4 · WEBM

  • 1
1 / 1 files
Required spoken video, limited to 120 seconds. The model page accepts MP4 and WebM. Ensure the people shown have agreed to appear speaking a translation in a voice like their own; overlength clips are refused without a charge. at least 1 characters
Defaults to English. Select among 178 base-language and regional-dialect options, including two English accent targets. Changing the target leaves the per-second rate unchanged.
Generation Modeaffects price
PrecisionSpeed
Use precision, the default, for its higher-quality lip-sync aim. Use speed for its faster-result aim at half the Precision rate; completion time is not guaranteed.
Safety checkerOff

SpicyAPI does not run a safety checker on this model. Your prompt and media go to the model as written, and the model itself decides what comes back.

Checking your account before enabling submission and pricing.

Output

Ready for a new run

Related models and next tasks

Missing the model you need?

TRANSLATE SPEECH AND SYNC LIPS

Give your talking clip a new language and matching lip movement.

Key numbers

Language options
178
Includes base languages, regional dialects and two English accent choices.
Source duration
120 S
Clips over this limit are refused before any charge.
Generation modes
2
Precision is the default; Speed has half its per-second rate.
Mouths re-synced
LIP SYNC
In our tests, mouth shapes changed to match the translated speech.
Result format
MP4
In our tests, the result used AAC audio and kept the source frame size.
Video retention
14 DAYS
Generated videos are deleted after this period.

Before you spend

Video translation

Consent comes first: use your own footage or obtain agreement from the people who will appear to speak the translation. The source must contain audible speech and last no more than 120 seconds. This page has no prompt, source-language field or subtitle option. Translated videos can differ in duration from their sources. Generated videos are retained for 14 days and then deleted.

What it does

PRECISION EXAMPLE

A cellar conversation translated into Spanish

This example pairs a two-person cellar scene with Spanish and Precision mode. In our tests, matching source and result moments showed different mouth shapes while faces, clothing and the scene remained the same.

SPEED EXAMPLE

Try Japanese with the same source recording

The Japanese example uses Speed mode for the cellar clip. Speed aims for a faster result and has half the per-second rate of Precision; the source duration still determines the billed seconds.

VOICE AND LANGUAGE

Language choices include regional dialects

The 178 options include Spanish (Mexico), French (Canada) and Chinese (Cantonese, Traditional). HeyGen says its voice cloning keeps the speaker’s tone when changing languages.

Made with HeyGen Video Translate

  • Spanish target, Precision mode: a two-person conversation in a candle-lit cellar.
    9s
  • Japanese target, Speed mode: another language choice for the same cellar source.
    9s
  • French (France) target with Speed selected for the night-market recording.
    6s
  • German target with Precision selected for the same night-market source.
    6s

Where it fits

  • Translate a course introduction

    Use your own short welcome video to introduce a course to Spanish-speaking learners, with translated speech and lip sync.

  • Adapt a product walkthrough

    With the presenter’s agreement, translate a spoken product demonstration into French for customers in another market.

  • Share a creator update abroad

    Translate your recorded announcement into Japanese so viewers can hear the update in their language.

Model reference

HeyGen Video Translate API reference

HeyGen Video Translate API specifications

Input video
Required video_url; MP4 or WebM with audible speech
Source limit
120 seconds; longer submissions refused without a charge
Translation choices
178 language and dialect options; English by default
Mode setting
precision by default, or speed
Voice approach
HeyGen says voice cloning preserves the speaker’s tone across languages
Measured output
In our tests: MP4, AAC audio, source frame size retained
Charge basis
Source seconds rounded upward; minimum 1, maximum 120

How to call the HeyGen Video Translate API

  1. Get agreement before translating

    Use a recording of yourself or people who have agreed to this translation. The first run on this page opens a consent notice you must accept.

  2. Provide a short spoken video

    Supply video_url for an MP4 or WebM with audible speech. The source must be no longer than 120 seconds.

  3. Select a language and mode

    Choose target_language, which defaults to English. Use Precision for its higher-quality lip-sync aim, or Speed for its faster-result aim and lower rate.

HeyGen Video Translate API questions

All HeyGen models3