Interesting progress for non-cloud service speech recognition using deeplearning/tensor flow.
https://hackaday.com/2018/01/17/speech-recognition-for-linux-gets-a-little-closer/
http://blog.mikeasoft.com/2017/12/30/speech-recognition-mozillas-deepspeech-gstreamer-and-ibus/
Looks Borg' worthy to me !
Nice
I can see where this could be useful for a nmber of droids such as Inmoov or R2D2 or even Work-E :-)
Yummy!
Fantastic, looks like a tensorflow model.. we should be able to load that tensorflow model with either DL4J or with the new java bindings for Tensorflow that have been added to MRL. Wow! This is absolutely fantastic! Though, I think the minimum requirements for CPU & Memory might have just increased a bit as a result :)
If I've got some free time, I'll start trying to poke around and see if I can get something worky with it! (Unless someone else gets to it first :) )
DeepSpeech
I was literally about to post about this, Grog. Guess you beat me to it ;)
I think it's in the best interests of MRL to add open source alternatives for the cloud based services, so we are no longer dependent on corporations that can change or monetize those APIs we depend on. Borging in DeepSpeech is definitely a step in the right direction :)
Wooiii, that could get us
Wooiii, that could get us free from Google speech recognition!
Very nice if that can be borged in!
More work for the Elves, but they enjoy it anyway, because new toys to play with is always fun.
:)
More work for the Elves
Hello Gael,
Maybe you could ask the Head elf when you see him in person :-)
Yeah we did talk about it.
Yeah we did talk about it. Briefly though, because we had SO many things to talk about!