Make every video speak your audience’s language

AI dubbing replaces the spoken audio in a video with a generated voice track in another language. Reap transcribes what is said, renders it in the target language, generates the dubbed speech, fits that speech to the flow of the original footage, and returns a video you can review and download. You do not write a script or record a voiceover.

Reap dubs into 80+ languages, offered as 143 target locales. Locales are counted separately because Spanish (Mexico) and Spanish (Spain) are different choices, which is why the locale number is the larger one. Coverage changes as languages are added, so the supported-languages reference and the get-dubbing-languages API endpoint are the authoritative lists.

Yes. Languages are selected as locales, so you can choose Spanish (Mexico) instead of Spanish (Spain), Portuguese (Brazil) instead of Portuguese (Portugal), or English (India) instead of English (United States). Arabic alone has more than a dozen regional variants. Selecting a locale sets the language and voice region; it does not rewrite culturally specific examples for you.

No. Reap generates a natural-sounding voice for the language and region you select. It does not clone or reproduce the original speaker's voice. The dubbed delivery is paced to the original video, but the voice itself will be a different one. Listen to the output before publishing.

No. Text-to-speech reads a script you supply. Dubbing starts from the video's own audio: Reap transcribes the speech, renders it in the target language, generates the voice in the target locale, and aligns the result to the original footage. The input is your video, not a script.

No, and we do not claim it. Reap uses lip-aware timing: the dubbed speech is paced and placed to stay aligned with the flow of the original footage rather than dropped on top of the original audio. Lip movements in the source video are not regenerated.

No. Dubbing replaces the spoken audio and nothing else. It does not generate or burn in subtitles. Captions and subtitles are a separate feature in the same Reap workflow, covering 100+ languages including romanized scripts such as Hinglish and Romanized Urdu, and you run that step yourself if you want them. Pairing a dubbed audio track with captions in the same language helps viewers who watch on mute.

Yes. Download the video from YouTube, upload it to Reap, and pick a source and target locale. YouTube dubbing works the same way as any other upload, and the dubbed file comes back for you to review and re-upload to your channel. Reap does not publish to YouTube on your behalf as part of dubbing.

No. Reap is AI dubbing software that runs in the browser, so there is nothing to install. The same dubbing is available over the REST API and to AI agents through Reap's MCP server, so you can dub video from the dashboard, from code, or from an agent.

The same limits apply in the dashboard and over the API: from 3 seconds up to 10 minutes, and a maximum of 2 GB. Reap accepts MP4 or MOV with clear audio. Processing is asynchronous, so poll for status or wait for the completion email. Dubbing is not a live or real-time feature. Current limits are in the dubbing walkthrough and the Create Dubbing API reference.

Yes. Dubbing is one step in Reap's workflow, not a separate product. The same source video can be clipped, dubbed, captioned, reframed to 9:16 or 1:1, styled with a Brand Template, and then exported or scheduled to your connected social accounts.

Yes. Upload a file, then POST to /automation/create-dubbing with an uploadId, a sourceLanguage, and a targetLanguage. The call returns a project you can poll with get-project-status, or you can configure a webhook and be notified when the project reaches a final state. The endpoint is rate limited to 10 requests per minute per API key. The same operation is available to AI agents as the dub_video tool on Reap's MCP server.

Yes. Automation removes the recording workflow, not your editorial judgement. Check people and company names, product terminology, technical vocabulary, regional pronunciation, tone, timing, and any on-screen text that still shows the original language before the video goes out.