Reviewing Alibaba's QoderVoice: A Full-Duplex Voice Agent for Desktop
AGI Hunt · wechat · 2026-08-01
The author provides an in-depth review of QoderVoice, a desktop real-time voice agent launched by Alibaba's Qoder. It resides as a floating bubble on the screen, featuring a full-duplex mode that allows users to interrupt it anytime while it executes tasks and reports progress. With screen-viewing permissions, it acts as a highly interactive AI co-worker.
Key Test Scenarios:
- Terminal Config: Modified tmux settings via voice to add a red border to the active pane, with the AI proactively optimizing background contrast.
- Software Guidance: By 'seeing' the screen, the AI accurately identified why a user was stuck in ScreenFlow (e.g., selecting an audio clip instead of video) and guided them step-by-step.
- Autonomous Operation: When asked how to install plugins, the AI took control of the mouse and keyboard to navigate the interface itself.
- Knowledge Management: Translated a browser article into Chinese and saved it to Obsidian; acted as an IELTS speaking coach while automatically logging vocabulary and notes locally.
Tech & Limitations:
Powered by Alibaba's Qwen-Audio-3.0-Realtime model, which ranks highly on Speech-to-Speech benchmarks. The author notes current limitations like fixed voice tone, lack of parallel sessions, and minor noise cancellation issues, but concludes that this bidirectional collaboration is a massive upgrade over unidirectional voice dictation.
More from coding & agent
- 6-Step Workflow for UI Design Using Lovable and Claude — damienghader · 2026-08-01
- Designing a Production-Grade MCP Server: 3 Principles for Unsupervised Agents — Capable-Necessary814 · 2026-08-01
- Managing AI Agents: The Reshaped Role of the New Engineering Manager — bendee983 · 2026-08-01
- Google Launches ADK Go 2.0 with Graph Workflows and Collaborative Agents — ZabihullahAtal · 2026-08-01
- Swarms v14 Released: Multi-Agent Framework Rebuilt, Claims 60x Faster Than LangGraph — KyeGomezB · 2026-08-01
- DeepSeek Harness Enters Closed Beta, Recruiting Agent Open-Source Developers — jiayuan_jy · 2026-08-01