Rhasspy is an open-source, offline voice assistant designed for private, local control of home automation systems. It targets savvy amateurs and advanced users seeking a customizable, internet-disconnected alternative to cloud-based assistants, offering seamless integration with platforms like Home Assistant and Node-RED.

How It Works

Rhasspy operates as a collection of independent services communicating via MQTT, adhering to a superset of the Hermes protocol. Users define voice commands using a template language, allowing for precise control. The system supports various speech recognition (ASR) and natural language understanding (NLU) engines, including Kaldi, Pocketsphinx, and DeepSpeech, enabling customization for different languages and vocabularies. Its modular design allows for easy extension or replacement of components.

Quick Start & Requirements

Installation typically involves choosing an installation method (e.g., Docker, binary) and accessing a web interface for profile configuration and training. Key requirements include compatible hardware and potentially specific Python versions, though the README does not detail specific versions or setup time. Links to a Getting Started Guide and a video series are provided.

Highlighted Details

• Supports multiple languages including English, German, Spanish, French, Italian, Dutch, Russian, Greek, Hindi, Mandarin, Vietnamese, Portuguese, Swedish, Catalan, Czech, and Polish, with various ASR/NLU engine options per language.
• Integrates with Home Assistant, Node-RED, Jeedom, and OpenHAB via MQTT, HTTP API, and Websockets.
• Offers a web interface for configuration, programming, and remote testing, with all functionality exposed via an HTTP API.
• Allows for custom voice commands with uncommon words or pronunciations, with phonetic assistance for new words.

Maintenance & Community

Community contributions are welcomed, with various ways to contribute listed. Users can seek help via the Rhasspy community site (Discourse) or GitHub issues.

Licensing & Compatibility

The project is open source with a permissive license, suitable for commercial use and integration with closed-source systems.

Limitations & Caveats

The README does not provide detailed system requirements, specific installation commands, or performance benchmarks. The development status of individual repositories is presented in a table without explicit pass/fail indicators or detailed issue counts.