Vosk4Unity is a module for the Unity Engine that provides a simple way to integrate speech recognition into your application. It is based on and powered by the Vosk speech recognition engine.
Vosk4Unity can be used with any Vosk-compatible speech recognition model and even provides an in-editor GUI for downloading and managing models. This allows for a very quick and easy setup.
The current development version only supports Windows x64 as the target build platform. Eventually, there will be two separate releases. One for targeting Windows 64-bit platforms and one for Windows x86 platforms. Any other target platforms, such as Linux, Android or macOS are currently not supported.
If you do need to build for 32-bit Windows clients, you can use the following work-around for now:
- Download the Windows 32-bit release of the official vosk-api here.
- Go into the folder 'Assets/lib' and replace the .dll files with the ones you downloaded
- Recompile