This project provides a cross-platform, real-time, offline speech recognition plugin for Unreal Engine, leveraging OpenAI's Whisper technology via whisper.cpp. It targets game developers and interactive application creators seeking to integrate voice command functionality directly into their Unreal Engine projects without relying on cloud services.

How It Works

The plugin integrates the whisper.cpp library, a C++ port of OpenAI's Whisper, enabling efficient, on-device speech-to-text processing. This approach allows for real-time transcription and command recognition directly within the Unreal Engine environment, bypassing the latency and cost associated with cloud-based APIs.

Quick Start & Requirements

• Install: Download the plugin from the Unreal Engine Marketplace (Fab version recommended) or compile from source.
• Prerequisites: Unreal Engine 4.26+ or 5.0+.
• Resources: Requires sufficient CPU resources for local model inference.

Highlighted Details

• Real-time, offline speech recognition.
• Cross-platform compatibility.
• Based on OpenAI Whisper and whisper.cpp.

Maintenance & Community

The open-source version is no longer maintained. The developer recommends the Fab version for updates and support. A Discord server is available for community interaction.

Licensing & Compatibility

The license is not explicitly stated in the provided text, but the project is based on Whisper (MIT License) and whisper.cpp (MIT License). Compatibility with commercial Unreal Engine projects is likely, but the lack of explicit licensing for the plugin itself warrants verification.

Limitations & Caveats

The open-source version is explicitly stated as no longer maintained and may contain bugs or lack features. The developer strongly recommends the Fab version for current functionality and support.