Develop
Develop
Select your platform

Voice SDK sample project

Updated: Apr 15, 2026
The Voice SDK sample project⁠ demonstrates how to use Voice SDK features in an Unreal Engine project. The project includes sample maps for voice commands, text-to-speech (TTS), and dictation.
Screengrab of the sample project

Prerequisites

Sample maps

The sample project includes two maps that demonstrate different Voice SDK features.

Voice commands and TTS

The VoiceSDKShapesDemo⁠ map demonstrates voice commands and text-to-speech. Use voice commands to change the color of objects in the scene. The map also includes a text input field as an alternative to voice input. For more information about voice commands, see Voice Command.

Dictation

The DictationDemo⁠ map demonstrates speech-to-text transcription using the WitDictationExperience actor. Speak into the microphone and see your speech transcribed in real time. For more information about dictation, see Dictation.

Getting started

To set up and run the sample project, follow the instructions in the GitHub repository README⁠. The setup process involves the following steps:
  1. Create a Wit.ai app using the backup file included in the repository.
  2. Download the sample project from GitHub.
  3. Run the install_VoiceSDK.bat script to install the Voice SDK plugin into the project.
  4. Open the .uproject file in Unreal Editor. The Voice SDK plugin compiles on first launch.
  5. Configure your Wit.ai access tokens in the ShapeConfiguration data asset, located at Content/VoiceSDK/Demo/Configuration/.
  6. Play the level to test Voice SDK features.
Note: For detailed instructions on each step, refer to the README in the sample project repository⁠.
Nav Logo
Virtual reality / Meta Horizon
Virtual reality / Meta Horizon
Virtual reality / Meta Horizon
English (US)
© 2026 Meta