MotionShine01 / Listen
Written out
Whisper, an open speech model, writes out what is said, with the time of each line.
Free tool
Audio translator, speech to any language.
Drop in a recording, a voice message or a video, pick the language spoken and the one you want, and get the translation line by line next to the original, with the times. Fix any line, save it as text or subtitles, and hear it read aloud by an AI voice in nine languages. Everything runs in this tab, so your recording is never uploaded.
Read it aloud
Spoken in English, Chinese, Spanish, French, German, Russian, Arabic, Hindi or Bangla. Other languages give text and subtitles.
Your recording is written out and translated in your browser. Nothing is uploaded.
How it works
Listen, translate, speak.
02 / Translate
Line by line
Chrome's built-in translator, or open Opus-MT models, translate each line. Edit anything before you save.
03 / Use it
Text, subtitles or voice
Save plain text or SRT subtitles with the original times, or have an AI voice read the translation as an MP3.
Questions
Good to know.
Is my recording uploaded?
No. The speech model, the translation and the voice all run in this tab. Models are downloaded once and kept by your browser.
Which languages?
About 40 languages can be picked. In Chrome 138 and later on a computer, Chrome's own translator handles most pairs. In other browsers, open Opus-MT models are used, which cover pairs with English directly and others through English. Reading aloud works in the nine languages our voices speak.
How good is the translation?
Good enough to follow a conversation or a talk, but machine translation can miss names, jokes and short lines. Read it before you send it to anyone, and fix lines in the boxes.
Can I translate a video with its subtitles on it?
Yes, with translate a video, which burns the translated subtitles into the picture. To lay a translated voice over your video, save the SRT here and use the AI voice over tool.
Can it translate live, as people talk?
No. It works on a recording. Record the conversation with the voice recorder first.