hermes-vtuber

What hermes-vtuber does

The repository description calls it a Hermes Agent integration for Open-LLM-VTuber. The README text is that of the upstream project, so it explains what Open-LLM-VTuber does rather than how the Hermes connection works. That project is a voice-interactive AI companion with real-time voice conversations, visual perception through camera, screen recording and screenshots, and a Live2D avatar, and it can run entirely offline on a local machine.

Described features include voice interruption without headphones, touch feedback on the avatar, Live2D expressions mapped to emotions, a transparent desktop pet mode, display of the AI's inner thoughts, proactive speaking, persistent chat logs, and text-to-speech translation so a chat in one language can be voiced in another. It runs on Windows, macOS and Linux as a web version or a desktop client. The upstream text notes that remote access from another machine needs https through a reverse proxy, because browser microphones require a secure context.

Key features

• Real-time voice conversation with a Live2D avatar
• Visual perception through camera, screen recording and screenshots
• Offline operation with local models, per the upstream README
• Desktop pet mode with transparent background and click-through
• Web version and desktop client on Windows, macOS and Linux
• Persistent chat logs and text-to-speech translation

When to use it

• Giving a Hermes agent a spoken voice and an animated avatar
• Running a local, offline voice companion on a desktop

Who it is for: Hobbyists and developers who want a voice-driven avatar front end for an AI agent.

How it fits with Hermes Agent

Described as a Hermes Agent integration for Open-LLM-VTuber, though the README does not explain how Hermes is connected.

Note: The README is the upstream Open-LLM-VTuber README and does not describe the Hermes-specific integration.

FAQ