A voice control setup for Home Assistant where wake word, speech recognition and spoken replies run on a local Jetson box. It can also bring HomeKit and Zigbee devices such as Aqara into one system. It suits homes and integrators who want voice control without sending audio to a cloud.
Homes and buildings
Results reported by the design's authors on their own test setups. Yours depend on the site, so we confirm them in a pilot.
The problem
Households want to control lights and appliances by voice without sending their speech to the cloud, and to bring HomeKit and Zigbee devices under one hub.
The flow
Deployment architecture
The devices on site, what carries their data, where the decision is made and where the result goes.
01SenseOn site
reSpeaker XVF380002ConnectCarries the data
03DecideRuns the model
reComputer J40121 per home
reComputer J30111 per home04Your systemsWhere results land
05ActWhat happens next
How it works
The hardware
Items we stock can go straight in the bag. Everything else we source and quote with the installation.
reComputer J4012fully local voice hub with room for a 7B LLMreComputer J3011lighter voice hub that forwards chat to an LLMreSpeaker XVF3800USB far-field mic array with echo cancellationBefore you install
Good to know
How we deliver it
Questions
Yes in the fully local setup. Speech and automations run on the LAN. Only the cloud LLM option needs internet.
The page lists English and Chinese only, so Indian languages would need testing or another speech model.
Yes. Devices bridged through Home Assistant can also be controlled from the Apple Home app and Siri.
Tell us the site and the problem. We come back with a pilot plan and a priced hardware list.
More in Acoustic & Audio AI
Based on a design published by Seeed Studio, written up by us. Symbiotes Supply is an independent reseller; product names belong to their makers.
