As the title says, I bought this when it first came out expecting all sorts of goodness and when I did have it up, it sucked. The sound is terrible, it’s not like i was going to use it for music but they advertise it as a whole home zoned audio device (I guess 3.5mm to a pre-amp, but still).
The “shoe” board has no ETA and you can’t pipe audio replies out to sonos speakers unless you use the 3.5mm out. Something voice PE and respesker boards can do effortlessly.
The new enclosure system is another $100 on top of the $100 dev board, so a $200 voice assistant that comes nowhere even near the quality of Amazon devices.
When issues are opened in the repo, they try and push you to discord, which surprise, surprise, isn’t indexable so all problems/solutions are locked behind discords system. They will close issues without any comments on open and being worked on feature requests (being worked on by community because they don’t want to implement things they promised) and then won’t answer simple questions asking if the feature has been merged or what the deal is.
So, all in all, I’ve got a $200 paperweight while my Alexa sits on top of it and absolutly owns it (not that I use Alexa much at all).
Is the top tier of self hosted voice assistants this? $200 sub-par experience and then needing to shell out god knows how much for the PoE “shoe” board (because their selling point of extra sensors on the dev board are useless for actual room readings, nevermind the ld2012 placement) AND THEN the supposed nexus AI hub.
That’s all to say its on hardware that isn’t special at all and fwict, the end user experience won’t be getting any better.
At least seeeed has the 4mic closed source XMOS, so we may get advanced DSP and such. You can replicate the FPH sat1 with a cheap seeeed respesker lite board and a full size esp32 s3 for a fraction of the cost and is what I am currently running.
It’s just sad that here we are dealing with the same subpar crap. Products rushed to market to ‘be the first’ is what I think is happening here.