MrCrispi
I was thrilled to see the new feature of "speech Recognition pausing when speaking" in a recent release, because I have quite a loud (amplified) voice on my robot and a lot of Mp3 sound triggers (over 150 in fact!) and it routinely "interupts itself" when speaking as something it is saying triggers a new Mp3 track to play - in fact it can get itself into infinite loops this way
However, I am not seeing any improvement at all and was wondering whether a) this perhaps isn't actually working yet or b) the fact i am using a sparkfun MP3 trigger to hold/play the MP3 clips means that this functionality improvement doesn't work. I know the problem is in the "listening while speaking" area because if I turn the speaker right down low the problem goes away - I then just can't hear the robot any more though, which kinda defeats the point!
Any insight or experiences of others gratefully accepted
Not sure how you are planning on doing this DJ, but it occurred to me today that if the mp3triggerplay command was given an extra variable (alongside baud rate etc.) which was how long to pause before the next track can play, then it would be easy to measure the length of each track in milliseconds and pause it for the requisite time for each track, rather than having one default pause for all tracks. If the default was set to zero, then all those folks with mp3 tracks already setup wouldn't be inconvenienced and it would be a great addition for all those of us with loud robot voices
Ahh, man! I hope this fix is figured out and added. This same problem has given me tons of stress. I also have worked very hard on getting my full size B9 LIS Robot to "chat" with me. It's an amazing thing to have him answer questions and make comments to things he "hears". In fact this is the main reason why I have built my robot around the ez-b robot board. It's frustrating for him to cut off in the middle of what he's saying to start another comment and like MrCrispi says it cansometimes go on forever if I don't stop it.
I've found some success with directional microphones and placing it as far away from the robot's speakers as possible. I've thought about building a sound box with only opening pointing away from the sound source. If this can be fixed or worked around through software that would be great. For me this has been the biggest (and only) disappointment with this otherwise outstanding platform.
I understand that it's a Windows "bug" and it really sucks. You would think MS would have done something about it. I wonder if Windows 8 will be any better and how this board and ARC will work with it.
Thanks and I CAN NOT wait , Dave Schulpius
Dave, are you using an mp3 trigger?
Because if you are using the speech synthesis, then it has been addressed. Please read the thread to see that it has been addressed for speech synthesis.
Yes DJ, I'm using the mp3 trigger. I've got a ton of voice files by the original actor who did the voice of the Lost in Space robot loaded on it. When it doesn't interrupt it's self, It's like having a conversation with the original robot from the old TV show. Very cool. When I read the other day that you were thinking of looking into way to keep the mic from listening while sound files played from the MP3 trigger I was jumping for joy!
I hope you can find the time and energy to address this somehow. It sounds like you and others that are smarter then me on this stff have some ideas on how to do it. Cant wait to see what you come up with.
Thanks, Dave Schulpius
Ill put something together for you
Thanks dj your the man as always
You are indeed "the man" - I can't wait...this will be the icing on the cake for me with this fantastic product & concept
Thank you! Thank You! Thank You!