HomeAssistant local voice processing
5 views
Skip to first unread message
Al Johnson
Mar 25, 2026, 6:44:13 PMMar 25
to derby-...@googlegroups.com
https://www.kunalganglani.com/blog/self-hosted-voice-assistant-home-assistant-2026-guide
So it seems you can do the basic structured commands (like Alexa used to
use) on a Pi 4 with 4GB now if you don't mind it taking a few seconds,
or with a low end CPU and 16GB to be comfortably quick. I think it's
OpenAI's Whisper speech-to-text that's made the lower end hardware
viable - this part used to need the sort of hardware that now makes the
local LLM possible (unless I misunderstood when I last looked into it!)
I might have to give this a go...
So it seems you can do the basic structured commands (like Alexa used to
use) on a Pi 4 with 4GB now if you don't mind it taking a few seconds,
or with a low end CPU and 16GB to be comfortably quick. I think it's
OpenAI's Whisper speech-to-text that's made the lower end hardware
viable - this part used to need the sort of hardware that now makes the
local LLM possible (unless I misunderstood when I last looked into it!)
I might have to give this a go...
Brendan Stafford
Mar 26, 2026, 5:10:52 AMMar 26
to derby-...@googlegroups.com
Yes, my understanding is that the minimum spec processor for it is the Intel N100 - which is what I bought for my new NAS
--
You received this message because you are subscribed to the Google Groups "Derby Makers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to derby-makers...@googlegroups.com.
To view this discussion, visit https://groups.google.com/d/msgid/derby-makers/c10ba893-03e3-4fb0-ad8f-bb4b1f4fe64d%40adj.org.uk.
Reply all
Reply to author
Forward
0 new messages