Robyn Inmoov Vlog 2


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
GroG's picture

Awesome Markus ! On Youtube,

Awesome Markus !

On Youtube, perhaps you should have a link to your 2 python scripts in github ?

Ive done similar telerobotics to another robot (not InMoov) - but watching the thing, and watching through a headset yourself gives me a "weird out-of-body feeling" - kind of fun .. unless it ends badly like it did a cyborg in "city of lost children" - where the victim gets to watch the strangler, strangle him through the murders eyes .. yikes !

The voice is being "recognized" then republished through (which Speech service) ?  
The recogniztion adds a substantial delay.  I was wondering if there would be interest in just relaying your voice directly over network ?  There is a TardosDsp library which can take your voice and realtime (noDelay) convert many aspects of it - pitch, speed, tone, depth - lots of effects if you want to sound instantaneously robotic.

Great installment ! .. already looking forward to the next ! 

Keep up the great work.

 

Ray.Edgley's picture

Neural Networks

Is it possible to use Neural networks such as Deeplearning4j or a MovidiusNCS to handel the voice rcognition locally?

Once trained, even a Raspberry Pi 3 can run the MovidiusNSC, Training just has to be done on a larger system.

The other downside of the Movidius NCS is it only supports Linux as the OS for now, there are two API, Python and C++.

But it would help to eliminate some of the voice recognition delay and also remove the requirement for the internet access for voice control....