Mycroft – Open-source voice assistant in trouble
kickstarter.com
kickstarter.com
TLDR: Mycroft is now out of money (reportedly due to ongoing patent litigation) and can no longer fulfill pledges. Though it's not clear if you can still pay for a full price unit?
> Which brings us to the present day. We will still be shipping all orders that are made through the Mycroft website, because these sales directly cover the costs of producing and shipping the products. However we do not have the funds to continue fulfilling rewards from this crowdfunding campaign, or to even continue meaningful operations.
If I would do this I would rather repurpose (and potentially resell) off the shelf mobile phones with a custom stand and rather get the software right first.
Technically I think we're getting close to a good offline system. Home Assistant solves the practical aspect of controlling systems and triggered routines - it really works well, at least on par with vendor software. Speech recognition models are usable and getting better. The interesting "glue" is parsing the commands into something reasonable and I think open source LLMs are likely to be the answer. See Open Assistant [0] for example. It doesn't solve the knowledge base issue (e.g. people seem to use Alexa either as a very fancy egg timer, or for querying facts), but it would probably allow for very natural interaction with the device and the model could just translate a query into a command that triggers an action.
My impression is that Amazon's approach (I have no knowledge of what goes on inside) has been to build fairly complicated subsystems to handle very specific queries. Each "Skill" has to be separately developed and if you ask something outside a fairly rigid grammar, the device has no idea.
That being said, shouldn't it be trivial to find prior art; prior software or articles already published on the internet that does what USPTO allowed Plaintiff to patent (2017) after searching only its own database?
(In my opinion, on a PC patents are vexatious.)
def on_voice_input(audio):
text = speech_to_text(audio)
tasks = parse_text(text) #
for task in tasks:
exec(task) # run locally, or
exec("ssh user@remote_host -- $task") # run remotely
Pretty sure there were voice command apps for Palm, Symbian, Blackberry, WinMo, etc; years before the date on the cited patents.
(And furthermore, such has certainly been possible on desktop computers (PCs) for many years prior to 2017.) E.g. Dragon Naturally Speaking (now Nuance) has been able run to custom commands `on_voice_input()` on a PC since the early 2000s at least IIRC)
Given voice commands prior art, to then run a remote command - over SSH, for example - is an obvious next step and thus not a patentable invention to this reasonable person.
Thus there is also prior art for running remote commands on remote devices (whether they be PCs, Macs, Linuxes, or mobile devices)