Hold a hotkey to record. Release it, and the recognized text appears at your cursor. Faster than typing, lighter than a dictation suite.
Download VIPEverything needed for voice input, without extra workflow or account friction.
Release the hotkey and text is usually inserted in under a second, with live subtitle preview while recording.
VIP pastes the result into the focused text field and restores your clipboard afterward.
Switch recognition language in settings: Mandarin, English, Japanese, Cantonese, Traditional Chinese, or Korean.
Bind Ctrl, Shift, Alt, Win, or Cmd with A-Z and F1-F12 keys.
A compact floating overlay shows recording state, volume feedback, and partial recognition text.
Optionally mute system audio during recording so speakers do not leak into the microphone.
Save recognition history locally, copy previous entries, and export everything to a text file.
Run quietly in the tray on Windows or as a LaunchAgent on macOS.
No installer framework or runtime dependency. The Windows build is only a few megabytes.
The default built-in server works out of the box. Advanced users can switch backend in settings.
Download the build for your platform and launch it. VIP appears in the system tray or menu bar.
The built-in backend needs no account. To use Volcano Engine instead, enter your App ID and Access Token in settings.
Default hotkey: Ctrl + Left Win on Windows, Ctrl + Cmd on macOS. Release to insert text.
Free and open source.
Use the built-in server immediately, or follow the guide to configure Volcano Engine for your own account.
View setup guide →