Towa Yoshida Vox Shift v1.0.44 Multilingual

Towa Yoshida Vox Shift is a standalone Windows utility for real-time speech translation and AI-generated voice output. Built for PCs with NVIDIA CUDA support, it captures speech from a chosen microphone or audio input, translates it, and produces spoken audio for live communication. It suits streamers, gamers, content creators, and music producers who need translated speech for broadcasts or other live audio workflows rather than a conventional VST effect.

Reference voice recording helps personalize the generated output, while noise cancellation supports cleaner microphone input. A monitoring output lets you check the resulting audio, and the built-in soundboard plays saved clips. Streamers also get OBS subtitle integration.

A setup wizard assists with device selection and audio routing, with model readiness checks to help prepare the required AI resources. Processing and storage are local by default, including recordings, translated text, subtitles, and settings; users may need to download model files separately. Routing translated speech into another application may require a virtual audio device such as VB-CABLE, which is not included with the Microsoft Store version.