Qwen Audio Agent Voice Runtime

Skill by ara.so — AI Agent Skills collection.

Overview

Qwen Audio Agent is a real-time voice runtime that enables AI agents to maintain continuous voice conversations while executing background tasks. Unlike traditional chatbots that pause during task execution, this runtime keeps agents "present" - listening, responding, and updating users about task progress naturally.

Core capabilities:

• Full-duplex real-time voice interaction with natural interruption support
• Parallel frontend conversation and background task execution
• Automatic task result integration back into ongoing conversations
• Multiple backend agent support (OpenCode, OpenClaw, Qoder, Kimi Code, etc.)
• Cross-session memory and user profiles
• Multiple interfaces: WebUI, Terminal TUI, macOS desktop app

Installation