Home Assistant 2023.11

This is very helpful, thanks. I am grateful for your explanations of the voice aspect - Amazon has kept all of this under wraps so much so that I didn't understand all the difficulties until I read your site just now. I can guess at the difficulties based on how often I have to repeat simple commands to Alexa, or "normal people bread" (my fam's word for non-grain, non-healthy bread) gets changed to "normal maple bread," things like that.

Practical question - are there plans to commercialize a turn-key Willow setup? Basically, I want an Alexa setup without the dependence on (or financial expectations from) Amazon. Key features:

* Few terminals, maximum reach. I want to be heard without having a cube every 10 feet. I'm okay shouting if need be. The ability to whisper is nice, too.

* Reminders, timers, alarms, schedules. Alexa, set egg timer for three minutes. Alexa, remind me six months from now at 8pm to change the furnace filter.

* Cloud-based lists: shopping shopping shopping. Alexa, add milk to the shopping list.

* Home automation: plugs, lightbulbs minimum. Thermostat nice to have.

* Routines. Alexa, turn on the living room lights. Alexa, fix the TV (occasionally the HDMI splitter freaks out and quits working; "fix the TV" turns off the power plug, waits five seconds, then turns it back on again.)

* Music: ability to play Spotify. Nice to have is "play this everywhere" capability.

* Basic math/language capabilities: Alexa, what is one point oh three raised to the 30th power (e.g., for calculating inflation over 30 years)? Alexa, define opprobrium.

* Other integrations: bluetooth, especially for speakers.

As for the other Alexa features - I only "reorder" things from Alexa. I never shop for new things, because shopping requires me to read reviews, see alternatives, who else bought what else, visit websites, and so on. I think Amazon missed the mark here - or rather, the product team promised senior management that the shopping experience would be easy, but it's not. Fail.

(If they charged me a nominal fee for Alexa, I would happily oblige. Sssh, don't tell them that.)

I also am itching to try a Raspberry Pi temperature probe, and somehow tie it into my outdoor pool heater.

I switched from Samsung's Smarthub to HASS - been really happy with the move. Took me a while to wrap my head around some of the manual configuration. Regardless, due to strong community support, HASS plays nice with devices that I could never get to work with Samsung's HA.

Yep, I'm relatively new to Home Assistant but love it. Combined with ESPHome, I've got most of the lights in my house controlled (via switch changes, not smart bulbs), and my all my air conditioners via the MHI-AC-Ctrl project[1,2] (which is fantastic - ordered the boards from JLC-PCB, parts from Ali-Express (AUD$2.50 ESP8266 boards), told ESPHome to program everything and plugged them into the ports and...it all just worked!).

Just got a Mitsubishi Electric (which is different to MHI) and a Daikin at my parents house to get online - but code for both exists and I suspect the MHI boards can be reprogrammed to fulfill the role without any hardware changes.

It's incredible that the future we wanted in like, 2010 can be had and had without selling your soul to the cloud: I can leave the house, and turn off all my lights and AC's while walking out the door, or get notified if I've left my garage open - without a single cloud provider dependency, all open-source.

EDIT: And yeah...I'm starting to eye anything which has a required cloud connection as "maybe I should just crack that open and see if we swap an ESP in..." - haven't done it yet.

[1] https://github.com/absalom-muc/MHI-AC-Ctrl/

[2] https://github.com/ginkage/MHI-AC-Ctrl-ESPHome

My pleasure!

If you have a lot of time and patience you can read my very long recent response[0] to a user asking about this where I go into arguably far too much detail. For context Wyoming is the fundamental protocol for the entire Home Assistant/ESPHome implementation.

In terms of commercialization, yes and no. Speaking of experience and background - the real genesis of Willow was three years ago when I was approached by a large publicly traded company in the healthcare space to implement a voice assistant for their use cases. The underlying and fundamental technology just wasn't there yet and it never got past PoC stage.

That failure really bothered me. Earlier this year everything seemed to really align and thus Willow was born. Our monetization strategy is where this all started three years ago - commercial and "enterprise" use cases. It's the only market I know and have experience with.

This is actually relevant to my comments about fundamental approach and architecture. Willow was designed for this market and their use cases which needless to say has substantially higher expectations than the more-or-less consumer/"tinkerer" market. I will NEVER monetize consumer/home users directly (again, I don't understand the market) but they get to benefit from our substantially better approach.

All of that said I'd certainly be interested in partnering with someone to bring Willow to the consumer market commercially. I've spent my entire career with the "sales cycle measured in months with a big customer for a lot of money" approach, not the "tons of consumers paying $5 a month" approach.

Everything you're describing is what we're currently focusing on. We feel we have the fundamentals of voice and audio pretty much nailed. Then the next question becomes "Great, I have performant and accurate speech in and out. What do I actually do with it?".

Taking the speech recognition transcripts and matching them with flexible grammar, functionality and application routing, etc is the next significant challenge. We have some of our own logic internally that we will be building out but at this early stage we heavily rely on the command processing and intent recognition from our supported command endpoints.

Nothing you're describing is impossible (of course) but Alexa (for example) has an eight year head start. It's going to take a while to reach feature parity with that. Our intention is to not only match Alexa functionality but also far surpass it by leveraging the entire open source ecosystem.

[0] - https://github.com/toverainc/willow-inference-server/discuss...