Open Voice OS

Open Voice OS, often shortened to OVOS, is a community driven, open source voice AI platform for building custom voice controlled interfaces across many kinds of devices. It emphasizes privacy and security and is designed so developers can assemble their own voice assistant rather than rely on a closed commercial one. The project began as MycroftOS and was renamed to avoid trademark issues. It now maintains its own core software while keeping its components compatible with the Mycroft ecosystem. Speech recognition and text to speech both use a plugin architecture, with online and local or offline options available, so users choose the engines that suit their privacy needs. Skills extend what the assistant can do, such as controlling smart home devices, playing music and radio, answering questions, and setting timers and reminders. Wake words can be customized. Supported hardware includes Raspberry Pi models, Mark I and Mark II devices, Linux desktops and laptops, and Docker containers. The site offers a prebuilt image for Raspberry Pi and embedded devices, plus an installer that supports Docker or a Python virtual environment for developers. Documentation, a blog, a contribution guide and community discussion channels are published on the site, and the project has received an NGI Zero Commons grant. Open Voice OS is free and released under the Apache License 2.0, permitting personal or commercial use, modification and redistribution. Donations are welcomed but entirely optional.