A Wyoming protocol ASR proxy that verifies speaker identity and isolates voice commands from background noise before forwarding audio to a downstream speech-to-text service. Designed for Home Assistant voice pipelines to prevent false activations from TVs, radios, and other people - and to deliver clean transcripts even in noisy environments.
docker proxy-server embeddings home-assistant speech-to-text speech-processing asr voice-assistant voice-activity-detection asr-services wyoming-protocol
-
Updated
Feb 25, 2026 - Python