AI Video Translator
AI Dubbing translates recorded video into speech in 13 target languages. Sign in, upload your file and listen to a free preview of up to 30 seconds.
Start dubbingPreview the first 30 seconds before you spend any credits.
Illustration
0:00Hola, bienvenidos a nuestro canal.
0:03Hoy les muestro cómo empezar.
0:06¡Vamos a ello!
Translate a video in four steps
- Upload an authorized video. Sign in and upload MP4, MOV or WebM up to 200 MiB and 10 minutes. The source language is detected from speech.
- Preview the voice. Choose from 13 target languages and listen to up to 30 seconds without spending credits. Preview does not include lip sync.
- Generate and review. Check the quote, generate the full dub, then review the source transcript alongside the translation. Edit translated sentences and review each edit's quote before regenerating.
- Download. Export the dubbed video, mixed audio or voice-only audio, plus separate source-language and translated SRT/VTT subtitle files.
Dubbing and subtitles from the same recording
Dubbing replaces recorded speech with a translated voice track. Subtitles let viewers read the source or translated script. Both are generated as part of dubbing here; there is no separate subtitle-only or transcription-only mode.
Check transcription, names, pronunciation and meaning before publishing. Subtitle downloads contain separate source and translated files, not bilingual lines in a single file. Caption styling controls are not offered.
Background, timing and speakers
- Background audio. Speech is separated and the background is mixed with translated speech. Separation can leave artifacts; listen to the result.
- Sentence timing. Translated speech is fitted to each sentence's slot and the silence before the next sentence. Short or overlapping lines need review; exact timing is not guaranteed.
- Several speakers. Detected speakers receive stock voices. Speaker renaming, reassignment and multi-speaker cloning are not offered.
- Optional lip sync. AI lip sync is designed for one front-facing speaker in video longer than 1 second and up to 210 seconds. It runs on the full dub, costs extra per language and does not guarantee perfect mouth alignment.
Supported languages
English, Simplified Chinese, Brazilian Portuguese, Latin American Spanish, French, German, Japanese, Korean, Vietnamese, Arabic (Modern Standard), Dutch, Polish and Hungarian.
Choose up to 5 target languages for a full dub. Each language is quoted and charged separately.
Choose a workflow
Translate your original channel file with the YouTube video translator, prepare a short authorized clip with TikTok translate, or start from a recording with the audio translator.
Pricing and permissions
Review the quote before generation. Full dubbing and optional lip sync use credits; failed languages release their reservations. See pricing for the current plans and credit packs.
Upload only recordings you own or have permission to use, including voices, music and images. AI output needs human review. This is recorded-file dubbing, not live interpretation. Read the terms.
From upload to dubbed video in four steps
No studio and no voice actors, and you stay in control of every sentence.
- 1
Upload your video
Upload recorded video or audio, up to 10 minutes and 200 MiB, and choose your target languages.
- 2
Preview 30 seconds free
Hear the first 30 seconds in your target language before you spend any credits.
- 3
Edit any sentence
Fix a translation and regenerate just that line; the rest of the dub stays as it is.
- 4
Export the dubbed video
Download the full video with the translated voice and subtitles.
Frequently asked questions
It converts recorded speech into translated audio and subtitle files. The pipeline detects the source language, translates sentences, generates speech and mixes it with the separated background.
Yes. You get a preview of the first 30 seconds without spending credits, then review the quote before generating the full dub. After the full dub, you can edit translated sentences; each edit has its own quote.
Yes, optional cloning can use a single identified speaker with their consent and at least 10 seconds of usable sample speech. Otherwise use stock voices. Voice similarity is not guaranteed.
MP4, MOV and WebM video, or MP3, WAV and M4A audio, up to 200 MiB and 10 minutes per file.
The dubbed video, the dubbed audio track on its own, and subtitles in SRT and VTT for both the original and the translated language.
Yes. Pick up to 5 target languages for the same upload; each language is billed separately by video length.