Image & Video

NeuralCompanion: Open-source local AI companion with realtime voice and avatars

Build your own AI companion on your hardware with voice, images, and avatars.

Deep Dive

NeuralCompanion (NC) is a new open-source project by Rakila & LAinol that reimagines the AI companion as a local-first, fully hackable desktop application. Unlike locked-down corporate assistants, NC gives users complete control by running entirely on personal hardware. It combines realtime voice chat, local LLMs (large language models), text-to-speech/ speech-to-text (TTS/STT), and image generation into one modular platform. The app also integrates with avatar engines such as VSeeFace, VAM, and VAM2, enabling experimental realtime avatar workflows. A built-in addon system and API-friendly design make it extensible for developers and artists.

The project is still experimental and acknowledges rough edges, but that's intentional — it invites contribution and customization. NC targets developers, tinkerers, AI enthusiasts, and anyone interested in local AI, creative tools, voice interfaces, or automation. The goal is to build a personal AI platform you can modify and shape yourself, not a black-box service. With its focus on privacy, modularity, and user agency, NeuralCompanion offers a glimpse into the future of truly personal AI companions that users own and control.

Key Points
  • Open-source, local-first AI companion that runs entirely on your own hardware without cloud dependencies.
  • Integrates realtime voice chat, local LLMs, TTS/STT, and image generation with a modular addon system.
  • Supports avatar engines like VSeeFace and VAM for experimental realtime avatar workflows.

Why It Matters

NeuralCompanion empowers users to build fully private, customizable AI companions without corporate restrictions.