Hello Everyone.

First, I’d like to thank everyone who contributes and maintains this site. It’s been a wealth of information for me while researching lately, and I really appreciate that.

I’ve been building and programming mobile robots most of my life. Since before there was a publicly available internet (sigh…I’m old. lol).

My current robot is Anna. She was developed on an electric wheelchair platform, and I’ve just recently moved her to a custom built chassis. She’s somewhat autonomous, using speech recognition and TTS, kinect camera for object recognition and navigation. I’ve also taught her to harass my wife’s cat. lol. She’s also remote-controllable through internet connected laptops and smartphones.

The next phase of my work is going to be giving her conversational abilities. I have two young neices that really want to be able to converse with her. I’m hoping that by integrating some chatbot technology that this will be possible.

The three most important goals for me (as far as the chatbot software) are : 1. Eventually, I’d like the interactions to be via voice, but I can handle that in c++.  2. I’d like to be able to read/write external system commands (ie: what are your ping sensors showing?) through the chat interface instead of WSRM, and 3. I’d like Anna to be able to “learn” facts or do lookups from local documents and external websites.

I’ve got a lot on my plate for this, I know. Be prepared for lots of questions from me! I’ve been doing serious research on Bruce’s ChatScript, and I really like it. I’m also trying to learn more about AIML 2.0. I’m still in the research phase, so at this point, I’m just trying to find the best option before I begin the nitty gritty building phase.

Thanks for having me!

John

Welcome, and interesting project you have there. I’d be interested to see a photo of your robot .
I’m a c++ programmer myself. We’ve recently had a discussion about linking up chatbots and windows speech recognition. It should certainly be possible to integrate a chatbot system to handle conversation, and “learning” to a degree, but the others can tell you much more about that.
The biggest difficulty I see is making speech recognition work reliably (for multiple voices), but on the other hand a chatbot would essentially only need to hear a few keywords right, and I imagine young children are easily pleased

Thanks for saying hello, and for the link, Don. I somehow missed that thread entirely in the last couple of weeks. I intend to post pictures of her soon. Before and after. Before was just a hideous electronics box mounted on the base of an electric wheelchair. I used an arduino to mimick the joystick voltages and voila: a very strong and controllable base.

In the last couple of months I built a new chasis, getting rid of the wheelchair frame. She stands about a yard tall, and about 18-20” square. With 4 decks for batteries, electronics, etc.  I’ll see what I can do about getting pics up soon.

John

Welcome John.
Two thoughts on different tangents:
1. I have noticed that speech recognition is much less effective with children than adults.  Maybe it’s just the way they tune it, but we have Chatty Addie on my ipad and it really struggles with my 5 year old.
2. Is there any research about combining vision with speech recognition? This would be a very advanced challenge, but just as humans are more likely to understand you if they can see you, so it could be for robots.

Hello OliverL, and thank you.

1. I agree about the recognition difficulties with children. For them, I will use typing as a fallback. The recognition is more for ME, when giving demos or commands, etc.  Extending it to others will, I’m sure, require fallback to the keyboard for input.

2. I’ll have to check into that research. I haven’t found any yet, but here’s a simple flow of how it could work. I’m using RoboRealm at the moment for object/face recognition. Let’s assume it recognizes ME. Now, with some C++ programming, it can load MY personal grammar file into SAPI for speech recognition. If, for instance, it recognized my wife’s image, then it would load HER grammar file. Interesting concept even though it would require some “training” for each user. Thanks for the idea…I’m going to chew on this for while. I like it. lol.

And I might be able to add some of the gesture recognition info from the Kinect SDK into the mix. Great! A new bunny trail to follow. I love it.

Thanks OliverL
John

Nice project John.  I say: lots of fun…

Thank you, Jan. I’ll be asking for more advice on Chat Script vs AIML when I get a little more research done. So far, I’m leaning heavily towards using Chat Script, but I also realize that I don’t know enough to make an informed decision yet.

I’m really happy with my mobile robot so far. The conversational ability will be a real nice touch, I think.

John,

Checkout these success stories with real world robots
and their amazing conversational abilities using RiveScript.

http://www.rivescript.com/blog/kirsle?skip=20

Along with some RiveScript source code samples for robots.

http://www.rivescript.com/yabb2/YaBB.pl?num=1350915838

There must be a reason RiveScript has been preferred
in robot projects.  RiveScript has been shown to work
with robots in the real world. So RiveScript is certainly
worth a look, and may be the de facto leader when it
comes to robots.

Thank you for the links, PlaNet. I appreciate it. I’ll check them out today.