Descrição
Aliyun Realtime Voice lets your UE5 characters listen, think, and speak in real time.
🎙️ Built for Alibaba Cloud Model Studio, this plugin connects realtime ASR, Qwen LLM streaming, and CosyVoice realtime TTS into one simple Blueprint-friendly Actor Component. Add the component, enter your own Alibaba Cloud API key, configure your models and voice, then start building realtime AI voice conversations.
✨ Simple Blueprint workflow
The plugin uses Unreal Engine's AudioCapture plugin for microphone input and exposes clear events for recognized speech, assistant text, TTS audio start, PCM audio chunks, audio finish, and errors. It is suitable for NPC dialog, AI assistants, virtual characters, and digital human projects.
🧑🎤 MetaHumanAudioToFace ready
Realtime PCM audio stream events can be connected to [MetaHumanAudioToFaceRuntime](https://www.fab.com/listings/7ba4ef64-5026-4c83-b3b1-bd4e1e0e5ff5), allowing generated speech to drive MetaHuman facial animation and lip-sync through the BeginAudioStream, FeedAudioStream, and EndAudioStream workflow.
🎚️ Configurable voices and models
Supports conversation history, system prompts, region selection, Workspace ID, model names, voice ID, volume, speech rate, pitch, microphone controls powered by AudioCapture, and mute-while-speaking behavior.
📘 Documentation:
🎧 CosyVoice Voice List:
Requires your own Alibaba Cloud Model Studio API key, internet access, and Unreal Engine's AudioCapture plugin for microphone input. Different CosyVoice models support different voice IDs, languages, and regions, so please check the official voice list when selecting a voice.