Audio files to translated speech

AI Dubbing translates recorded audio into speech in 13 target languages, with a free preview of up to 30 seconds before you generate the full dub.

Start dubbing

Preview the first 30 seconds before you spend any credits.

What this audio translator does

AI Dubbing turns recorded speech into a spoken translation, with subtitles for the source and target language. Use it for podcasts, interviews, voice notes or narration that you want listeners to hear in another language.

How to translate an audio file

Upload your recording — MP3, WAV or M4A, up to 200 MiB and 10 minutes.
Choose a target language and preview. The source language is detected automatically. Listen to a free preview of up to 30 seconds without spending credits.
Generate the full dub and review it. Correct names, terms or phrasing sentence by sentence; only edited sentences are regenerated.
Download the result. Get the audio mixed with the background, the voice-only audio track, and separate source-language and translated SRT/VTT subtitles. An audio upload does not produce a video.

Speakers and voices

Cloning is optional: select it and confirm you are the voice owner or have their permission. It requires a single identified speaker and at least 10 seconds of usable sample speech; multiple speakers or insufficient speech fall back to stock voices. Stock voices are assigned by detected speaker, with reuse when the voice catalogue is exhausted. Listen to the preview and review the full result before publishing.

Background and timing

Speech is separated from the background, then translated speech is mixed over the separated background. Separation is not guaranteed to remove every trace of the original voice. Timing adaptation attempts to fit translated speech to the source timeline; short or overlapping lines still need review.

Languages

The 13 target languages are English, Simplified Chinese, Brazilian Portuguese, Latin American Spanish, French, German, Japanese, Korean, Vietnamese, Modern Standard Arabic, Dutch, Polish and Hungarian.

Other recordings

Use the Spanish to English audio translator for that language direction, the Spanish to English video translator for a video, or TikTok Translate for creator clips.

Pricing and permissions

Dubbing is billed by source duration and target languages. Check the credit quote before starting the full dub; failed target languages are refunded. See pricing.

Upload only recordings you own or have permission to use, including the voices, music and images. AI transcription, translation and pronunciation can contain mistakes; review the result before publishing. See the terms. This tool processes recorded files, not live conversations.

Frequently asked questions

How do I translate an audio file to English?

Upload a recording and choose English as the target language. The source language is detected automatically. Listen to the free preview, generate the full dub, review and edit the translated sentences, then download.

Does it keep the original speaker's voice?

Yes, with optional voice cloning and the voice owner's consent. Cloning requires a single identified speaker and at least 10 seconds of usable sample speech; otherwise the result uses stock voices. Voice similarity is not guaranteed.

Do I also get a transcript?

Yes. Download separate source-language and translated subtitles in SRT and VTT. These are individual files for each language, not a single file with bilingual lines.

Which audio formats are supported?

MP3, WAV and M4A, up to 200 MiB and 10 minutes per file. MP4, MOV and WebM video uploads are also supported.

Is background music kept?

Yes. The separated background is mixed with the translated speech. Separation can leave artifacts; listen before publishing. A voice-only audio download is also available.