Runtime Speech Recognizer is an [[MIT License]]d offline speech recognition system for [[Unreal Engine|Unreal]] written in [[C++]].
- [Website](https://www.unrealengine.com/marketplace/en-US/product/runtime-speech-recognizer)
- [GitHub](https://github.com/gtreshchev/RuntimeSpeechRecognizer)
- AlternativeTo
> Runtime Speech Recognizer is an open-source plugin that enables real-time, offline speech recognition. Based on Whisper OpenAI technology, particularly [whisper.cpp](https://github.com/openai/whisper) library, and supports multiple language models pre-selected in the plugin's settings.
# Notability
# Philosophy
# OS Support
# Features
- English-only and multilingual models available, with multilingual supporting 100 languages
- Different model sizes available (from 75 Mb to 2.9 Gb)
- Automatic download of language models in the Editor
- Optional translation of recognized speech to English
- No static libraries or external dependencies
- Cross-platform compatibility
>[!WARNING]
> The Marketplace version of the plugin may have issues with staging the language model asset due to an engine problem in locating the asset specified in **DirectoriesToAlwaysCook** (**Additional Asset Directories to Cook** in the editor) for the modules located in the engine folder, which could lead to an inability to use the plugin in the packaged build. This Marketplace-specific issue is currently under investigation. If you encounter such an issue, please consider downloading the plugin directly from [**GitHub**](https://redirect.epicgames.com/?redirectTo=https://github.com/gtreshchev/RuntimeSpeechRecognizer).
>
> Please be aware that in the 4.27 engine version, there are errors in speech recognition, particularly in streaming mode. Therefore, it is highly recommended to use the >=5.0 engine version.
# References