Make your Unity characters hear, think, and talk — using real voice AI. Locally. No cloud.
UnityNeuroSpeech is an open-source framework for creating fully voice-interactive AI agents inside Unity.
It connects:
- ๐ง Whisper (STT) – converts your speech into text
- ๐ฌ Ollama (LLM) – generates smart responses
- ๐ฃ๏ธ XTTS (TTS) – speaks back with custom voice + emotions
All locally. All offline.
No subscriptions, no accounts, no OpenAI API keys.
- ๐ฎ AI characters that understand your voice and reply in real time
- ๐ฟ NPCs with personality and memory
- ๐งช Experiments in AI conversation and narrative design
- ๐น๏ธ Voice-driven gameplay mechanics
- ๐ค Interactive bots with humanlike voice responses
| Feature | Description |
|---|---|
| ๐๏ธ Voice Input | Uses whisper.unity for accurate speech-to-text |
| ๐ง AI Brain (LLM) | Easily connect to any local model via Ollama |
| ๐ฃ๏ธ Custom TTS | Supports any voice with Coqui XTTS |
| ๐ Emotions | Emotion tags (<happy>, <sad>, etc.) parsed automatically from LLM |
| ๐ฌ Actions | Action tags (<turn_off_lights>, <play_cutscene_123>, etc.) also parsed automatically from LLM |
| ๐๏ธ Agent API | Subscribe to events like BeforeTTS() to monitor your agents |
| ๐ History Saving in JSON | Save dialog history between player and LLM in JSON with/without AES encryption |
| ๐ ๏ธ Editor Tools | Create, manage and customize agents inside Unity Editor easily |
| ๐งฑ No Cloud | All models and voice run locally on your machine |
| ๐ Multilingual | Works with over 15+ languages, including English, Russian, Chinese, etc. |
| ๐ Multiple Voices and Languages for Multiple Agents | Each Agent can have each voice file for any available laguage |
| โก High Performance | Uses UniTask instead of Coroutines and Tasks for optimal performance. |
| ๐ง Full Build Support | Full compatibility with both Mono and IL2CPP scripting backends. |
- ๐ง OllamaSharp
- ๐ค whisper.unity
- โก UniTask
- ๐ง Coqui XTTS
- ๐ฅ๏ธ UV
- ๐ค Unity 6
| Scripting backend | Windows | Other platforms |
|---|---|---|
| Mono | โ | โ (not planned) |
| IL2CPP | โ | โ (not planned) |
See UnityNeuroSpeech official documentation.
UnityNeuroSpeech was created by HardCodeDev — solo dev from Russia.
UnityNeuroSpeech is licensed under the MIT License. For other Licenses, see Licenses.
