Audio translator: translate a recording and voice it

An online tool for translating audio recordings: upload a podcast, an interview, a lecture or a voice message, and the tool recognizes the speech, translates it and voices it in the language you choose. It reads 100 languages and speaks 32. Besides the audio you get a timestamped transcript and subtitles, and the music and background can stay where they were.

How to translate an audio recording

1

Upload an audio file — MP3, WAV, M4A, OGG, FLAC, AAC or OPUS.

2

Set the language of the recording or leave detection on, then choose the target language.

3

Pick a voice: in-browser, studio, or the speaker’s own.

4

Press “Translate and voice” and wait for the translated transcript.

5

Take the result: audio, TXT or SRT subtitles.

Translate a podcast, an interview or a lecture and hear it in another language

Audio translation comes with Premium: recognition on our server, studio voices and the speaker’s own cloned voice.
In-browser voices
Draft quality, synthesized on your device
Studio voices
Synthesized on our server, spends tokens
The speaker’s voice
Keeps each speaker’s own timbre
Speech is separated from the backing track, and the translation sits on top of the original.
Drop an audio file here or click to choose
MP3, WAV, M4A, OGG, FLAC, AAC, OPUS ·

Press ▶ on a line to jump to that point in the recording.

Translation and speech are produced by AI — check anything important before publishing.

What the audio translator does

100 languages in, 32 out
Speech is recognized in 100 languages, and the translation can be voiced in 32 — from English and Spanish to Vietnamese, Ukrainian and Chinese.
Music and background survive
A separate model splits speech from the backing track, so the translated voice sits on the original music instead of replacing it with silence.
A different voice per participant
The tool tells up to three speakers apart and gives each their own voice, so an interview does not collapse into one narrator.
Translation that respects timing
Lines are translated with the length of the original in mind and then fitted to their slot, so the translation does not drift away from the recording.
Transcript and subtitles
Along with the audio you take the text with timecodes: TXT with the translation, SRT in the target language and SRT in the original.
The speaker’s own voice
A separate lane carries each speaker’s timbre into the other language, so the recording still sounds like them.

What people use it for

  • Release a podcast in a second language while keeping the music stings.
  • Understand an interview or a call recorded in a language you do not speak.
  • Translate a lecture or a webinar for students from another country.
  • Make sense of a voice message sent in someone else’s language.
  • Get subtitles for audio in two languages at once.

How it differs from a text translator

A text translator works on a string, but a recording is also timing, several voices and the sound around the speech. Here the translation is tied to timecodes, lines are fitted to their duration, speakers are told apart and the music stays — what comes out is a recording you can listen to, not a list of phrases.

Frequently asked questions

Which formats does the tool accept?

MP3, WAV, M4A, OGG, FLAC, AAC and OPUS. If you have a video, use the Video dubbing tool instead — there the translated track goes straight back into the video file.

How accurate is the translation?

Speech is recognized by Whisper and translated by a language model that watches the length of each line. On a clean recording quality is high, but noise, a strong accent and overlapping voices lower it — check anything important against the transcript.

Can the speaker’s voice be preserved?

Yes, that is a separate lane: each participant’s timbre is carried into the target language. It covers ten languages and spends tokens.

What happens to the music in the recording?

With “Keep music and background” on, a separate model splits speech from the backing track and the translated voice is laid over the original music.

How long a recording can be translated?

Up to 60 minutes per run. A trial run is limited to five minutes; full length comes with Premium.

Do I get text, not just audio?

Yes. The timestamped transcript is shown on the page, and you can download it as TXT with the translation, SRT in the target language and SRT in the original.

You May Also Like

📝 Audio to Text 🗒️ Dialogue analysis and conversation protocol 🎤 Karaoke Maker 📢 Megaphone Effect 🎙️ Online Voice Recorder 🎚️ Re-Dub 🎤 Record Vocals 🎙️ Speech to Text

How do you rate this tool?

Thank you for your rating!
Want to share more? Leave a comment!
Thank you! Your comment will appear after moderation.
Who is this tool for?
🎙️ Podcasters 🎬 Video Creators ✨ Content Creators 🎓 Students 📚 Educators & Teachers
Published Author: