Runtime Text To Speech

Cross-platform, offline text-to-speech synthesis for Unreal Engine. Generate natural-sounding speech with 900+ voices across 40 languages - now featuring Kokoro studio-quality voice models.

UE 4.27 - 5.5
Blueprints & C++
All platforms supported
40 languages, 900+ voices

Offline Text-to-Speech Made Simple

Runtime Text To Speech provides real-time, offline, and cross-platform text-to-speech synthesis with no internet connection required. The plugin supports 40 languages with over 900 voices and 160+ voice qualities, including the new Kokoro studio-quality voice models for exceptional output quality.

Completely Offline

Works without an internet connection across all platforms

Extensive Voice Options

40 languages, 114 voice models, and 172 voice qualities

Studio-Quality Voices

Featuring 53 high-quality Kokoro voice models across 7 languages

Seamless Integration

Easy to use with Blueprint and C++ support

Video Tutorial
Full Blueprint support with easy-to-use nodes

Key Features

Model Management

Download and manage voice models directly within the editor interface.

Download and preview voice models
Download, preview, and manage voice models in the editor

Multi-Language Support

40 languages supported with multiple voice models for each language.

English (US & UK)
German
Spanish
French
Chinese
Russian
Portuguese
Hindi
Polish
Italian
And 30 more languages...

Multiple Speakers

Many voice models support multiple speakers, significantly increasing voice variety - for example, English LibriTTS includes over 900 different speakers.

Get speaker count from model
Check how many speakers are available in a voice model

Technical Details

Voice Model Architecture

Piper TTS Engine
Kokoro Voice Models
ONNX Runtime
PCM Audio Output (Float32)

Voice Model Types

Standard Voice Models

Basic voice models with good quality for general use

Multi-Speaker Models

Voice models with multiple speaker options (up to 900+ in some cases)

Emotional Voice Models

Models with emotional variations for expressive speech

Kokoro Studio-Quality Models

High-quality models with exceptional naturalness and clarity

Platform Support

Windows
Mac
Linux
Android
iOS
Meta Quest

Powerful Integrations

Runtime Text To Speech works seamlessly with other plugins to create complete speech and audio solutions for your Unreal Engine projects.

Runtime Audio Importer

Process the synthesized speech audio data for playback, saving, or further manipulation with advanced audio processing capabilities.

Learn more

Runtime Speech Recognizer

Create two-way voice communication by combining speech recognition and text-to-speech for interactive conversational experiences.

Learn more

Runtime MetaHuman Lip Sync

Animate character lip movements in real-time using the synthesized speech for immersive character dialogues and interactions.

Learn more

Runtime AI Chatbot Integrator

Connect with advanced AI chatbots to generate dynamic text content that can be converted to speech for AI-driven character interactions.

Learn more

TTS Options: Offline vs API-based

Runtime Text To Speech provides offline, on-device speech generation with no internet connection required – ideal for games that need to work offline or with privacy constraints. Runtime AI Chatbot Integrator connects to cloud services like OpenAI and ElevenLabs for state-of-the-art AI chat and TTS capabilities when an internet connection is available.

Documentation & Support

Get started quickly with our detailed documentation and receive support through multiple channels. From basic usage to advanced techniques, we're here to help you succeed.

Comprehensive Documentation

Step-by-step guides for all features

Tutorials

Visual guides for common use cases

Discord Community

Get real-time help from developers and users

Email Support

Direct contact for custom development needs

Blueprint example

Demo Project

Try our demo project to experience the capabilities of Runtime Text To Speech firsthand. The demo showcases various voice models and features.

Demo GIF
Demo interface showing text-to-speech synthesis with different voice models

Demo Features

  • Selection of different voice models
  • Text input for custom speech synthesis
  • Speaker selection for multi-speaker models
  • Audio playback of synthesized speech
  • Blueprint implementation examples

Supported Languages

The plugin supports 40 languages with multiple voice models and qualities for each, including the new Kokoro studio-quality voice models.

Major Languages

  • English (United States) 19 models, 43 qualities
  • English (British) 10 models, 19 qualities
  • German (Deutsch) 8 models, 10 qualities
  • French (Français) 7 models, 8 qualities
  • Spanish (Español) 8 models, 10 qualities
  • Chinese (简体中文) 2 models, 10 qualities

European Languages

  • Russian (Русский) 4 models, 4 qualities
  • Italian (Italiano) 2 models, 2 qualities
  • Polish (Polski) 4 models, 4 qualities
  • Dutch (Nederlands) 5 models, 7 qualities
  • Ukrainian (Українська) 2 models, 2 qualities
  • Turkish (Türkçe) 3 models, 3 qualities

Additional Languages

  • Portuguese (Português) 4 models, 6 qualities
  • Hindi (हिन्दी) 1 model, 4 qualities
  • Korean (한국어) 1 model, 1 quality
  • Vietnamese (Tiếng Việt) 3 models, 3 qualities
  • Catalan (Català) 2 models, 3 qualities
  • And 25 more languages... Various models

Kokoro Studio-Quality Models

The plugin now includes 53 high-quality Kokoro models across 7 languages: English (US), English (UK), Simplified Chinese, Spanish, Portuguese, Hindi, and French. These models represent some of the highest-quality open-source TTS solutions available today.

Try Kokoro Voices Online

Add Text-to-Speech Capabilities to Your Project

Generate natural-sounding speech from text directly in your application with no internet connection required. Create voice-enabled experiences across all platforms with 40 languages and hundreds of voices.