Hello everyone,
I’ve been teasing this for several days, but here it is, it’s now live and I can announce it:
I’m going part-time on Gladys!!! 
I wrote an article on the blog to summarize all this =>
https://gladysproject.com/fr/article/grande-nouvelle-gladys
Don’t hesitate to reply here if you have any questions/comments! 
As I told you on Twitter, I wish you the best! 
So, I’ve become a Patron! 
But how does Discord work? Because I’m downloading the Mac client.
I don’t think you put the server name anywhere…
On the Patreon page, you should normally have a button that says « Join this Discord »!
I can’t add you manually, you need to do it via Patreon to be in the right roles!
Done too, once again thank you for this project! 
Personally, I don’t see it either 
Oh bother! It was supposed to be on the « thank you! » page
But maybe there’s another place…
I personally received an email for Discord
Me too but nothing more, it allowed Patreon on Discord but I haven’t seen anything more.
I’m discovering Discord…
Indeed, I spoke too soon
Same here! Once authorized, the server does not appear in my list..
So I found this help in case it can help you!
https://patreon.zendesk.com/hc/en-us/articles/212052266-How-do-I-receive-my-Discord-role-
I succeeded , just log out of Discord on Patreon and log back in on the Settings page! 12|690x206
Well done @spenceur! Thanks!
I looked into DeepSpeech, I saw that some people had asked about running it on a Raspberry Pi, it’s not for now but it could come with some work =)
Ah, you mean could the project one day host a DeepSpeech for Gladys users? Why not, clearly yes one day!
Oh! I was thinking it’s crazy that no one is talking about it!
Good news, the version of deepspeech is much better. It works on Raspberry Pi 4, and apparently extremely fast!
It’s also available on Rhasspy ;D
The most complicated part will be the integration
No Docker / Python images & dependencies. Version with/without CUDA
I’ll leave it to you to figure out
HS: I take advantage of following the subject to thank you for the nickname @VonOx
Finally, I suspect!! But I like it ^^
In this case, not a big problem, just need to create one (but rhasspy has it integrated directly )
And I read that they switched to a lite version of tensorflow that allows for really better performance, like 22 times less RAM impact, things like that
Damn, I got caught 
Oh, I see, I was convinced that rhasppy was just RPI (that it was an OS), didn’t realize there was Docker.
