qwen-audio-agent
A realtime voice runtime that keeps Agents talking, working, and present. Real-time Voice Runtime for AI Agents
What is qwen-audio-agent?
A realtime voice runtime that keeps Agents talking, working, and present. Real-time Voice Runtime for AI Agents
Overview of qwen-audio-agent
What it is
Qwen Audio Agent is a realtime voice runtime for AI agents. It is designed to keep agents talking, working, and present during conversations. The runtime supports full-duplex realtime voice interaction, natural interruption, and sustained multi-turn conversation. It integrates voice frontends and backend agents, allowing tasks to run in the background while the conversation continues. The project is open-sourced under the Apache-2.0 license.
Key features
- Full-duplex realtime voice interaction, natural interruption, and sustained multi-turn conversation - Replaceable realtime voice frontends, with cloud services and local deployment options - One-click integration with your preferred Agent, reusing its model configuration, tools, MCP, Skills, and authentication - Frontend conversation and background tasks run in parallel; ask about progress or cancel at any time - Create multiple independent tasks executed asynchronously by the backend Agent, with continuous status tracking - Task results automatically return to the current conversation, supporting follow-up questions and modifications - WebUI, terminal TUI, and desktop floating orb (macOS / Windows / Linux)
- Long-term per-user personalization and cross-session memory
Use cases
The README mentions examples such as Office and Smart Cockpit. The news section also lists customer service, smart cockpit, digital human, and other examples. The runtime is suitable for scenarios where an agent needs to maintain presence while performing background tasks.
Getting started
Requires Node.js 22.22.2+ or 24.15.0+, npm 10+. One-click install (recommended): bash npm install -g qwen-audio-agent Create your config and fill in the API Key: bash qwenaudio config dotenv QWEN_AUDIO_REALTIME_API_KEY=your-key # Voice frontend model: optional, defaults to Qwen Audio 3.1 Realtime Plus QWEN_AUDIO_REALTIME_MODEL=qwen-audio-3.1-realtime-plus # Backend Agent: optional, leave empty or set to none for frontend-only mode AGENT_PROTOCOL=openclaw # Backend model: optional; explicit values use standard ACP, empty reuses Agent config QWEN_AUDIO_AGENT_BACKEND_MODEL=qwen3.7-max Before starting, create a key from the Bailian API Key page.
License and pricing
The license is Apache-2.0. The project is open source with no stated pricing.
Based on the project's public README and repository metadata.