This tutorial demonstrates how to implement offline realtime text-to-speech in Unreal Engine using the Runtime Text To Speech plugin.
You’ll learn how to:
- Download and manage voice models for 40+ languages
- Synthesize speech from text without internet connectivity
- Play synthesized audio using Runtime Audio Importer
- Implement both regular and streaming speech synthesis
- Work with multi-speaker voice models
The plugin provides complete offline text-to-speech synthesis with 115+ voice models across 40 languages, including high-quality Kokoro models. It’s perfect for accessible interfaces, NPC dialogue, voice-driven tutorials, localization, and procedurally generated content.
Prerequisites: Runtime Text To Speech and Runtime Audio Importer plugins.
https://dev.epicgames.com/community/learning/tutorials/a6VD/high-quality-offline-real-time-text-to-speech-in-unreal-engine