Local-first default
The default configuration is designed not to send your audio or translated text to VoxShift operator servers.
Local / Privacy-first Voice Translation
Voice translation designed to stay on your PC by default.
In its default configuration, VoxShift (both the Steam and Microsoft Store versions) does not send audio data, reference voice, translated text, or caption content to VoxShift operator servers. Speech recognition, translation, and voice generation run on your PC, and the models needed for first use are included in the package. It is not advertised as fully offline, because purchase, license checks, and optional model downloads use the internet.
Related use cases
local voice translationprivacy-first voice translatorprivate voice translatorlocal AI voice translatorvoice translation on PC
Local model cache
The models needed for first use come with the package, and optional models are stored locally too. VoxShift checks your PC and picks a matching setup.
High-impact use caution
Use care in work, medical, legal, financial, or other situations where mistranslation matters.
Fit Check
Where VoxShift fits, and where it does not.
Good fit
- You want translated speech output for Discord, game VC, or OBS streams
- You need more than subtitles and want your voice chat partners to hear translated speech
- You want to test GPU, VRAM, audio devices, and latency with the free version first
Use caution
- You need zero-latency simultaneous interpretation or perfect speaker separation
- You need high-stakes medical, legal, financial, or emergency translation
- You plan to use another person's voice without clear consent
- You want Live voice translation on an AMD Radeon, Intel GPU, or no-GPU PC (setup and diagnostics only)
Setup Flow
Check this flow before you buy.
Live voice translation requires an NVIDIA CUDA GPU (GTX 16 / RTX 20 series or newer) with 4 GB VRAM or more, and results also depend on your audio devices and routing. Test the free version (Steam Demo or Microsoft Store free trial) on your own PC before purchasing. It limits each Live session to 10 minutes, translates into English only, uses a fixed light model set, and allows up to 3 soundboard items.
- 1Install the Steam Demo or Microsoft Store free trial
- 2Check the setup and estimated VRAM in the Performance tab
- 3Check audio devices
- 4Set routing if needed
- 5Review privacy and terms
FAQ
Common questions
Does VoxShift send my audio to its servers?+
By default, in both the Steam and Microsoft Store versions, audio and translated text are not sent to VoxShift operator servers. Recognition, translation, and voice generation run on your PC.
Is it fully offline?+
No. Purchase and license checks, updates, and optional model downloads need internet access. The models needed for first use are included, and recognition, translation, and voice generation run locally.
Does it bundle third-party drivers?+
The Steam version includes the official VB-CABLE installer from VB-Audio as an optional install. The Microsoft Store version does not bundle it and points you to the official VB-Audio page instead.
Start with the free version.
Voice translation is highly dependent on your PC. Before buying, check your PC, devices, and latency with the free version. Steam and Microsoft Store sell separate products, each a one-time purchase.