BROWSER EXTENSION

Translate any browser tab — and hear it, not read it.

The Voxis extension captures the audio of the tab you are on, translates it, and speaks it back in your language while the video keeps playing — with live captions underneath. 79 languages, in Chrome and Edge.

What it does

The extension asks the browser for the audio of your active tab, streams it to a real-time translation model, and plays a spoken translation back through your speakers a couple of seconds behind the original. The tab’s own audio is automatically turned down while the translated voice speaks, so the translation sits on top of the original rather than fighting it. Captions for both the source and the translation appear in the popup as the session runs, and accumulate into a transcript you can scroll back through.

It works on whatever the tab is playing: a YouTube video, a Twitch stream, a recorded lecture, a news broadcast, a meeting running in the browser. There is nothing to route and no virtual audio device to install. The extension only ever receives the audio the browser hands it, and only while a session is running — when you stop, it is not listening.

Two ways to run it

The same extension works either on VoxisLive minutes or entirely on your own API key. Own-key mode is covered by the one-time BYOK licence — the same $99.99 unlock as the Windows app, bought once for both — so it is worth being precise about what each mode means.

Two ways to run it
 Account modeBYOK mode
Where the translation key comes fromVoxisLive issues one per sessionYou paste your own Google Gemini or Alibaba Qwen key
What voxislive.com receivesYour sign-in, plan quota and minutes used — never your audioA licence check when you open the extension and when a session starts — your account token, nothing more. Never your audio, never your key
Where your key livesNever stored in the browser; issued per sessionOn your device only
CostDraws on the same minute balance as the Windows app; a new account starts with 15 free minutesOne-time BYOK licence, $99.99, covering the Windows app and the extension; after that you pay only your own API provider
SetupSign inSign in, unlock BYOK, paste your key

Both modes send the tab’s audio to the translation model — that is inherent to how real-time translation works. In neither mode does that audio pass through VoxisLive. What differs is the data path: in BYOK mode we are out of it entirely, and the one thing we are asked is whether your licence is valid. Full detail on the extension privacy page.

One key, everywhere

The BYOK licence is a single one-time purchase — $99.99 — and it covers both the Windows app and this extension. Buy it once and you can run on your own API key on your PC and in your browser, with no further per-minute charge from us.

Being plain about what changed: own-key mode in the extension used to be free and ungated, and it is now part of that licence. There is no grandfathering — a key saved in the extension before this change does not unlock it on its own. If you would rather not deal with API keys at all, nothing about account mode changes: it still runs on prepaid minutes and still starts with 15 free ones.

The free tier in the browser

The daily free tier works here too: a new account gets 15 free minutes of the real voice, and after that 10 free minutes every day, the same as the Windows app. What differs is the voice. On Windows and Linux the app ships its own neural voices; a browser extension has nowhere to run those, so it speaks with the voices your browser already has. On Windows that covers 18 of the 79 languages in Chrome and 16 in Edge — the two Edge is missing are Hindi and Indonesian — and the exact list depends on your browser and on the speech voices installed in your operating system. A language with no voice on your machine still translates; it arrives as live captions instead. The extension tells you before you press Start when the language you picked is one of them. The full breakdown is on the languages page. Prefer more spoken languages, or want to translate games and desktop apps? That is the Windows app.

What it will not do

A comparison page that only lists strengths is not worth reading, and neither is a product page.

  • It translates the tab, not you. The extension dubs incoming audio. It does not translate your microphone, so it will not speak for you in a call — that is the desktop app’s Meeting mode.
  • It only hears the browser. Games, desktop apps and anything outside a tab are out of reach by design; a browser extension can only be handed a browser tab’s audio. The Windows app captures the whole system.
  • Some protected streams stay silent. DRM-protected video on certain streaming services will not share its audio with any extension. The desktop app has no such limit, because it captures what the sound card is playing rather than asking the page.
  • It is a few seconds behind. A spoken translation cannot be instant — that is the same span a human interpreter needs to hear a thought before rendering it. If you need sub-second text, you want subtitles, not a voice.

Privacy

Audio is captured only while a translation session is running, and recordings are not retained once the translation has been delivered. The extension does not track browsing history, clicks, keystrokes, scrolling or location, and it does not collect health or financial data. In BYOK mode the only thing it sends us is a licence check — when you open the extension, and again when a session starts — carrying your account token and nothing about what you are translating. Your audio and your API key never reach VoxisLive in either mode. The full policy is on the extension privacy page.

Install

Add it from the Chrome Web Store — it runs in Chrome and in Edge. Open it on any tab with Ctrl+Shift+1, pick a target language, and press start. Pricing for account mode is the same prepaid minute balance as the desktop app; see pricing.

FAQ

Common questions

01What does the Voxis extension actually do?
It captures the audio of your active browser tab, translates it in real time, and speaks the translation back through your speakers in your language while the tab keeps playing — with live captions for both the original and the translation. It supports 79 target languages.
02Do I need a VoxisLive account to use it?
Yes — both modes need one. Account mode draws on the same prepaid minute balance as the Windows app, starting with 15 free minutes. BYOK mode runs on your own Google Gemini or Alibaba Qwen key and needs the one-time BYOK licence ($99.99), which covers the Windows app and the extension together. Even then your key and your audio stay out of our reach: when you open the extension and when a session starts, it asks voxislive.com only whether your licence is valid.
03Does the 10 free minutes a day work in the extension?
Yes. A new account gets 15 free minutes of the real voice, and after that 10 free minutes every day, the same as the Windows app. The difference is the voice: the extension speaks with the voices your browser already has, which on Windows covers 18 of the 79 languages in Chrome and 16 in Edge. A language your browser has no voice for still translates and arrives as live captions, and the extension says so before you press Start.
04Does it work on Netflix and other streaming services?
It works on anything the browser will share — YouTube, Twitch, news sites, recorded lectures, browser-based meetings. DRM-protected video on some streaming services refuses to hand its audio to any extension, and there is no way around that from inside the browser. The Windows app is not affected, because it captures the audio your sound card is playing rather than asking the page for it.
05Is my audio stored anywhere?
No. Audio is streamed for translation and is not retained on our servers once the translation has been delivered, and it is only captured while a session is actually running. In account mode VoxisLive receives your sign-in, quota and minutes used — never the audio itself. In BYOK mode the only thing VoxisLive receives is a licence check — when you open the extension and when a session starts — never the audio, and nothing about the session itself.
15 free minutes to start · 79 languages

Hear every language, in real time.

Runs in Chrome and Edge. Nothing to route, no virtual audio device, no bot in your call.