Speech-to-speech, not subtitles
VoxisLive treats a foreign podcast the way a native simultaneous interpreter would: it listens to the speaker, stays a few seconds behind, and plays a continuous spoken translation in your target language while the episode runs. You keep your eyes free and simply listen, instead of racing to read captions.
Because the output is spoken audio, it fits the way podcasts are actually consumed — on a walk, while cooking, or in the background while you work. If you still want a written record, every session is saved as a searchable transcript you can export afterwards.
Driverless capture from any app
VoxisLive captures whatever audio your Windows PC is playing using WASAPI process-loopback, so it does not care which app produces the sound. Podcasts streaming in Spotify, the Apple Podcasts web player, Pocket Casts on the desktop, a browser tab, or a YouTube episode all play through the same system audio mix.
There is no file to import and no URL to paste — VoxisLive works on live audio only. Start a session, press play on the episode, and it translates as the sound streams.
No virtual audio cable required
Unlike tools that make you route sound through a virtual device, VoxisLive reads the system audio mix directly. There is no VB-CABLE, no virtual audio device, and no driver install. Install it from the Microsoft Store with no reboot and start listening.
It also excludes its own translated voice from what it captures, so it never re-translates the audio it just spoke. This driverless approach is the core difference from setups that force you to configure routing before you hear a single word.
79 languages, your choice per show
VoxisLive supports 79 target languages. You pick the language you want to hear the podcast in, and the cloud model translates the incoming speech into a spoken voice in that language.
You can change the target language between sessions to suit whatever show you are listening to, so a Spanish interview and a Japanese tech podcast are equally at home.
Save, search, and export the transcript
Captions are not the main output, but every session is stored as a searchable transcript. When an episode ends, you can export it to TXT, SRT, or VTT — handy for pulling quotes, building show notes, or keeping citations.
For video and game audio, a one-way Video/Game mode ducks the original track under the translation so the spoken interpretation stays clearly on top.
Prepaid minutes or an open-source BYOK build
VoxisLive runs on prepaid minutes, so you only pay for the time you actually translate — no subscription lock-in. See current rates and bundles on the pricing page.
If you would rather bring your own keys, an open-source BYOK build is available on GitHub. Either way, you can install VoxisLive from the Microsoft Store and start translating podcasts in minutes.