State of the project

  Today is Tuesday Feb. 13 2017. Fred-X is still inert (sort of). all of his servos are working individually. The issue is that the Aacer laptop computer which controls him via mrl will not yet run more than one at a time.

  It seems like the voice command part is the problem. When I open "Start_InMoov" it appears to boot up properly and it says good morning to me but soon after that it does not react to vocal input. Something here is just not copacetic. I will get it to work eventualy.

  Printed a right side worm gear actuator overnight. It looks MUCH better than its predicesor. I may have to reprint the left one as well. Currently in the process of printing one of the BIG gears for the belly. That should be the last part, unless I reprint the left side worm. 2 days and 13 hours to do the big gear with no infill. 14 hours more for the worm. 

  I should be working on the software issue while that is going on, but will be out of pocket for a couple of days getting an eye fixed. As to the voice thing, I think I will run a power cord over to Fred and light off his Kinect. Maybe it is looking for THAT microphone.



Comment viewing options

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

Hello Mike! I've had some

Hello Mike!

I've had some issues with the speech recognition myself before. For whatever reason, Chrome blocks microphone access sometimes, even after I've hit allow access. You might want to check that. As for the Kinect, it requires special configuration to use its microphone. Try this for configuring it on Windows, but I doubt it will solve your problem. If speech recognition seems to be working (in the webgui, you'll see the microphone pulse and words appear above it as you speak), then I would open up the ProgramAB service (should already be running after Start_InMoov, can't recall what the service name is) and try entering commands manually. Just type them as if you were speaking, the input goes into the robot's "brain", and the speech recognizer is supposed to push the recognized text to that service. If the robot responds when entering commands manually, there's a communication problem between the speech recognition and ProgramAB. If it still doesn't respond when entering manually, send a noWorky (sends the log of MRL) so we can figure out what's going on.

Hope we can fix your issues, I'm excited to see your creation working!