# AI Dubbing > AI dubbing swaps the speech in a video for a translation in any of 33 languages, spoken in a clone of each original voice. Kenerate AI's dubbing runs on ElevenLabs: upload a video or audio file up to 15 minutes with up to 32 speakers, and download the dubbed MP4 or MP3. Lip sync is a separate step. URL: https://kenerateai.com/ai-dubbing Publisher: Kenerate AI (https://kenerateai.com) Last updated: 6 October 2026 ## Key facts - Target languages: 33 - Spoken languages: 57 + auto-detect - Longest file: 15 min - Speakers: Up to 32 - Voices: Cloned from each speaker - Lips: Separate lip sync step - AI Dubbing (ElevenLabs): Video or audio, up to 15 min → 33 languages, each speaker's voice cloned. Re-voicing speech in another language. - Lip sync (VEED, Sync Labs, ByteDance, WaveSpeed): The dubbed video → Mouth matched to the new language. On-camera faces after dubbing. - Auto Subtitles (VEED): Video up to 4 min → Captions burned in, in the spoken language. Viewers with the sound off. - Translate audio (ElevenLabs): MP3, WAV, M4A or a mic recording → Audio in the original voices, 33 languages. Podcasts, voice notes, interviews. ## Steps 1. Open AI Dubbing — Sign in with a verified email; new accounts get starter credits to try it. 2. Add a video or audio file — MP4 or MOV up to 500 MB, or audio up to 50 MB, up to 15 minutes. In audio mode you can record from the mic. 3. Pick the language to dub into — One of 33 target languages. Leave the spoken language on Auto-detect unless detection misses. 4. Set speakers and background — Speakers on Auto or 1–32; switch on Drop background audio for voices only. 5. Compare and download — Flip between the original and the dub on the result card, then download. Add lip sync if a face is on camera. ## In depth ### From one voice to another language The audio is split into speech and background, the speech is transcribed and translated, and each speaker is re-voiced in the new language with a clone of their voice, then mixed back with the background [1]. ElevenLabs says its dubbing aims to keep each speaker's emotion, timing and tone [1]. ### Dubbing vs lip sync Dubbing works on sound; lip sync works on the picture. Traditional film dubbing matches the translated lines to mouth movements by hand [2]. With AI, it takes two steps: dub the audio, then let a lip sync engine redraw the mouth on the dubbed video. ### What Kenerate's AI Dubbing doesn't do Knowing the limits saves a wasted run. - No lip sync inside dubbing: it is a separate tool. - No script editor: translation and voices are automatic. - No link import: upload the file itself. - One target language per run; run again for another. - No subtitle file: captions are a separate, burned-in tool. - Files over 15 minutes must be trimmed first. References: [1] ElevenLabs — Dubbing docs: https://elevenlabs.io/docs/capabilities/dubbing [2] Dubbing — Wikipedia: https://en.wikipedia.org/wiki/Dubbing ## FAQ Q: What is the best AI dubbing tool? A: It depends on the job. HeyGen suits presenter videos that need lip-synced translation, ElevenLabs is voice-first with a long language list, and Rask AI handles long videos with a transcript editor. Kenerate's AI Dubbing suits quick dubs of files up to 15 minutes in 33 languages, with lip sync and subtitles in the same studio. Q: Is AI dubbing free? A: You can try it without paying: new accounts get starter credits that work on AI Dubbing, and a verified email is needed. After that it is priced by the length of the source file, billed by the second up to 15 minutes. If a dub fails at the provider, the charge is refunded automatically. Q: Does AI dubbing keep my voice? A: Yes. ElevenLabs re-voices each speaker with a clone of their own voice, so a two-person interview comes back as two distinct voices in the new language. There is no stock-voice picker in the dubbing tool; for a script read by a chosen voice, use the Voice Studio instead. Q: Does AI dubbing match the lips? A: No. Dubbing changes the audio and leaves the picture as it is. For an on-camera speaker, run the dubbed video through lip sync, which offers seven video engines including VEED Lipsync 2 and Sync Lipsync 3, so the mouth follows the new words. Q: How many speakers can it dub? A: Up to 32 in one file. Leave Speakers on Auto and the number is detected, or set it yourself when people talk over each other, which helps the tool keep each voice separate. Q: What files can I dub, and how long? A: Video as MP4 or MOV up to 500 MB, or audio as MP3, WAV or M4A up to 50 MB, both up to 15 minutes. Video comes back as an MP4 with the new speech; audio comes back as an MP3. Longer files must be trimmed first. Q: What is the difference between dubbing, voice-over and subtitles? A: Dubbing replaces each speaker's lines with translated speech in their voice. A voice-over lays one narrator over the original. Subtitles leave the audio alone and show translated text. AI dubbing lets people watch without reading. ## Related pages - AI video translator and dubbing: https://kenerateai.com/ai-video-translator-and-dubbing - AI lip sync: https://kenerateai.com/ai-lipsync-generator - Translate audio: https://kenerateai.com/translate-audio - English to Spanish: https://kenerateai.com/english-to-spanish-voice-translation ## Sources - Kenerate AI Dubbing (tool): https://kenerateai.com/app/tool/dubbing - ElevenLabs — Dubbing docs: https://elevenlabs.io/docs/capabilities/dubbing - HeyGen — AI video translator, checked 6 October 2026: https://www.heygen.com/translate - ElevenLabs — Dubbing, checked 6 October 2026: https://elevenlabs.io/dubbing-studio - Rask AI — home page, checked 6 October 2026: https://www.rask.ai/ - Dubbing — Wikipedia: https://en.wikipedia.org/wiki/Dubbing