Open Voice OS (OVOS) is a community-driven, open-source voice AI platform for creating custom voice-controlled interfaces across devices. It features natural language processing (NLP), a customizable user interface, and a focus on privacy and security.
Key Features
- Community Driven: Powered by global developers from Linux and FOSS communities.
- Multi-platform: Supports embedded headless devices, single-board computers with screens, and DIY smart speakers.
- Experimental at Core: Provides a platform for experimental features before upstreaming to other Linux-based open-source voice assistant projects.
- DIY: Open to all developers and contributors wanting to support specific devices or platforms.
Installation
OVOS can be installed via Python virtual environment (for developers) or Docker (for stability). A one-line installer script is provided. Prebuilt images are available for Raspberry Pi and Mycroft Mark II devices.
Hardware Support
Raspberry Pi (3B, 3B+, 4, Zero 2), Mycroft Mark I and Mark II, Linux desktops/laptops, and Docker containers.
Offline Capability
ovos-core does not require a backend internet server to operate. Default speech-to-text (STT) requires internet, but offline STT options (e.g., Vosk) are available. Some skills work entirely offline.