You people are awesome! But we need more

Ok,

So here’s me…just trying to use ‘your’ AI for the first time… I would really like to replace ‘your’ with ‘our’… I really do. So here goes my first impressions and what ‘we’ need to improve…

Ok, I might be biased, but I really like what you are doing! I really like the ‘opensource’, I really like the ‘not sharing my personal shit’, I really do…but I think this is only a viable alternative…if we kick ass!

We are really in the right direction of ‘kicking ass’, but we are not really there yet…

I might be a ‘stand-alone-case’, but I like to think I’m not. I’m not tooo tech-savvy, I do know what I would want, and I think I do know what the majority wants (am I too blunt?) I think that you, this community is really on the right track, but also…you’re losing momentum on ‘the big guns’. I just tried briefly to add ‘MyCroft’ into my home…and as much as I would like to call MyCroft my buddy…I can’t just yet. Sorry, yes, I know a lot of you are really working on this on your spare time, I think ‘we’ could do better…‘we’ really should…

I really think ‘we’ should look at our strengths…

  1. Being able to easily change the ‘wake-up-word’, this is really…really important to a lot of us!!..this should be easy!!

  2. Being able to incorporate us into our ‘smart’-homes, this is really…really important to a lot of us!!..this should be easy!! (I tried, I really tried with Home-assistant, but really…come on, I can ask to turn on the ‘waterpump’ *friendly-name?! but if I ask the state of the ‘waterpump’, it says it doesn’t know ‘of’…really?!

  3. I personally really…really…like the design of Mycroft V1…is there any way to get V1 with V2 internals? Is there any way we could print V1 ourselves and use arduino and some extra’s acquired at mycroft.ai to make V1 ourselves? (with V2 functionality…by adding a Pi 4b ourselves?

  4. Is any real machine-learning in the works? If I ask Mycroft ‘could you repeat that’…it responds ‘that’…this in itself is hilarious, but also really sad… I would think in the 21st century…‘we’ could do better…

  5. I really would love a cost-effective solution to speak to ‘MyCroft’ in all parts of my house…just like the competition…I add a device in each room…and ‘boom’…I can talk there…

  6. I really love how this initiative is set up, I really love that it is open source and I really am very humble to all that are working on this…I just really hope ‘we’ could work towards something that could really make a change and kick all of these ‘commercial-privacy-selling-corporations’-ass :wink:

My 2 cents…

Tony

1 Like

Edit the mycroft.conf file and change the phonemes (http://www.speech.cs.cmu.edu/cgi-bin/cmudict) and word as you like.
Precise is a bit more involved but has better results.

That’s a function of interfacing two disparate projects no matter how you cut it. It’s always being looked at and worked on.

A pi 4 would mostly fit in to place. Probably better to find a different mounting for it, there’s a lot of internal cabling/boards to make that all work. Might even be better to use two small lcd’s instead for eyes as well.

Precise, mimic2, participation with common voice…

Not quite here yet, but always looking for people to help with that.

Thanks for coming along.

Prob cheapest is a Pi3A+ with 2x I2S Mic with software AEC as for many uses its essential but hardware is extremely costly.

As posted here.

Sort of dissagree about changing wake word as do you call your dog a cat because you prefer the name?
The load to provide phonetic custom based wake words is huge compared to specific trained models and now that server TPU hourly charging is so cheap fragmenting accuracy for a single phonetic customisable model rather than a choice of specific trained models is definately the wrong direction.
Satelites can be extremely lean and making the KWS lite has huge impacts on cost in this mode.

Unfortunately the importance of collating datasets is hardly mentioned as if they where available easily accesable with accompanying tools models would come in droves.
Supplying models without the datasets is as closed as suppling binaries without code.

Mycroft need to look over there shoulder as there are alternatives that in implementation have probably overtaken them.

https://linto.ai/

If they had a community portal it would probably be a no brainer as there direction with many aspects is superior, but seem business orientated.

Mycroft in a many distributed form is fat and expensive but hopefully we will see a more partitioned form soon that is official.

There is also the RockPiS which is a Pi3 perf 64bit Soc with built in audio and DSP VAD for $15 that prob needs exploring.

The skills section and community in Mycroft is still awesome but voice AI has rapidly changed, presuming with the new CEO and the new vacancies much is going to change as currently the hardware proposed is very hard to justify with what is available elsewhere.

Pi3A+, 2x I2S mics or USB stereo ADC soundcard, 40 watt amp & Speaker is approx $60 and about as cheap as I can make it.
But the hardware that seems to be proposed in terms of cost for a number of units to be littered around the home is definately an eye roller.