The Sat1 hardware won’t necessarily unlock that feature, but with a little hacking it should be possible to build what you’re describing today, I think? Perhaps you could use the multi-wake-word and multi-pipeline mapping feature and have one of the wake word/pipelines fire up a UDP stream (you’d have to custom-build a UDP streaming ESP component) targeting the correct STT endpoint for your application.
Just thinking out loud.
Oh I see. So this looks like it may be doable. I assume you will be using a highly modified version of voice kit. Or a replacement.
Should actually be more of a slightly modified version of voice-kit, actually. But it’s a bit hard to be 100% sure because everything is happening so fast on both projects. I’ll make a point to catch up with the HA voice guys here really soon to ensure we’re on the same path. Our FutureProofHomes/Satellite1-ESPHome repo should go live really soon.
@FutureProofHomes maybe it would be a good idea for you to create your repo as a downstream fork of esphome’s voice-kit repository then just add your own patches and modifications to yor forked repo as that way it should be easier to keep up with the HA voice guys upstream repo and even simpler to submit patches to them as upstream? See → GitHub - esphome/voice-kit
Not a bad idea at all. We’re working on getting our Github Actions Build & Release strategy dialed in so our ESPWebTools and releases are automated… once our repo stabilizes I’ll see what the viability of a downstream fork might look like depending on the deltas in our repo and firmwares. Thanks for thinking out loud! I like it.
Seems like a few boards are coming, this is another one. More based on the concept of a multi-sensor and local voice assistant…
I have an ESP32-S3-BOX-3, how do I set it up to have the 3 wake words? Is it possible yet?
How do we go about getting microwakeword v2 onto a box3-s3? Currently im using this:
substitutions:
name: esp32-s3-box-3-6405bc
friendly_name: Kitchen ESP Box
micro_wake_word_model: alexa
packages:
esphome.voice-assistant: github://esphome/wake-word-voice-assistants/esp32-s3-box-3/esp32-s3-box-3.yaml@main //<-- this guy
esphome:
name: ${name}
name_add_mac_suffix: false
friendly_name: ${friendly_name}
api:
encryption:
key:
I created a Wyoming Satellite using Pi Zero and 2Mic HAT but unable to get the timers or delayed commands to function properly. It does use the wyoming 1.5.4 version which supports timers. I am also using Open AI Conversation based conversation agent in my pipeline. Am I missing something?
I have listed every command I ran and the steps I followed in this in-depth guide:
I’d really appreciate if somebody could help me figure out what’s wrong with this!
Until the Satellite1 are available (again), what other ESP32+XU316-based boards are available at the moment for us to play with?
The ones I know about already:
(Edited to add projects mentioned in comments below)
FYI, much more on the ReSpeaker-Lite board in this other community thread:
Also check out ”Koala Satellite” 3D printer enclosure + off-the-shelf hardware for ReSpeaker-Lite:
Still hoping someone will design updated Onju Voice style PCB boards for the Google Home Mini and Google Nest Mini and release as open-source hardware, see request and related discussion:
…and that it be released under a non-restrictive open-source hardware license, read:
And
For reference, “Onju Voice” is a custom PCB (based on ESP32) designed to be a drop-in replacement to the original Google Nest Mini so that you can reuse rest of Google Home smart speaker hardware. Enable anyone to order single or batches of PCBs directly from PCBWay.
The downside is that the current variant does not feature an XMOS chips and is not open-source:
