Willow

Willow is a voice assistant for people who want to control home automation through hardware and speech processing they own. It provides an alternative to Amazon Echo or Google Home on Espressif's ESP32-S3-BOX hardware family.

• Offline commands: Recognize a configured set of voice commands entirely on the device, including on a network without internet access.
• Speech transcription: Use Willow Inference Server on your own server to recognize speech beyond the configured commands, or send audio to Tovera's hosted inference service.
• Home automation: Send recognized speech to Home Assistant, openHAB, or other applications that accept it through REST APIs or HTTP POST endpoints.
• Room-distance voice control: Choose a wake word such as "Hi ESP" or "Alexa" and speak from across the room; noise suppression and acoustic echo cancellation help with voice capture, and speech detection identifies when you've finished talking.

Willow is open source under Apache 2.0. On-device command recognition keeps that processing on the hardware; self-hosted transcription uses your own inference server, while Tovera's hosted service processes audio externally. Willow also supports the ESP32-S3-BOX-3's color display and touchscreen. Optional audio compression reduces traffic on busy 2.4 GHz Wi-Fi networks.