Currently, I am teaching myself the Processing language. Other than that, I do not have any programming experience. My plan is to build a robot that can hold basic conversations with people, but I am not sure where to start. Would it be easier to create a chatbot through a web host and download it to robot? Or, would it be better to create a chatbot on my computer? What would be a good place to start?

Hi, Zach, and welcome to chatbots.org.

I’m just curious as to which “processing language” you’re referring to (there are literally dozens, if not hundreds), and I certainly don’t want to discourage you from your desired goal, but what you intend is extremely ambitious, considering the skill and experience level that you’ve said that you have. the way you’ve described things, are we to assume that you wish to create, or already have, an actual, physical robot that you wish to integrate conversational software with? and if so, what sort of size are we looking at? If the robot is too small, the computer hardware required to utilize chatbot software AND TTS/Speech recognition software would be difficult to put in, due to the exceedingly cramped quarters. Plus, depending on any other functionality that you’ll want to have in place for your robot, other conflicts and/or obstacles may arise, as well. There’s a lot to consider here, so perhaps a much more detailed outline is in order, so that we can best assist you.

The language I am referring to is the one from http://processing.org/. I understand that the project is ambitious, but I am willing to invest the time to make the project work. My time horizon to finish the project is over a period of years, since I am at the beginning of my learning curve. My robot will have to run off a laptop or a tablet, as I will need a lot of processing power. Think of a talking “Eddie” robot from Parallax, combined with the follow me function of the Kinect sensor, yet able to have a conversation. I will be working on pieces of the robot a little at a time, and I am in the early planning stages. Hopefully, this will give you an idea of what I am trying to accomplish.

Yes, that’s a much clearer outline, thank you.

Given that you intend to use at least a tablet to run things, then you should have no trouble with putting together something that will meet your requirements (relatively speaking, that is ). If I were working on a project of this sort, the biggest challenge for myself would be in building the robot, so I won’t be able to advise you in that capacity very much, but the rest of it is something that I know a thing or two about, so here’s my 0.02¢ worth:

Unless you’re planning on giving the robot a WiFi connection that allows for continuous contact with the internet during operation, you’ll need conversational software that’s fully self-contained. You have several options here, ranging from something like Kari, which is “virtual Girlfriend” software, has a very shallow learning curve, and already has a built-in female voice (but has it’s share of “negatives”, too), to something like ChatScript, coupled with the talking GUI that I wrote for it, which is very flexible, already integrates TTS and Speech recognition, and allows for using a male or a even a “computer” voice, but has a very steep learning curve. Given that your time-line for this project spans years, there’s no real rush to make a “written in stone” decision, and we can discuss the advantages and disadvantages of each option, if you like.

ChatScript combined with your GUI seems like the best route to take in the long run. I downloaded ChatScript to see if I could manage it, and I think it will be fun to try and learn.  It seems that the negatives to ChatScript will be its steep learning curve, but what are the negatives of Kari?

There are, to my thinking, at least two disadvantages to using Kari. The first is that Kari, being designed as a “virtual girlfriend”, and mainly for a more ““adult oriented” individual, has certain pre-defined behavioral patterns built in that are somewhat hard to get rid of. In effect, “she” wants to be… um… “Intimate” much of the time, and often says so, in a way that is NOT family friendly. The second, and probably most important disadvantage is that, without built-in speech recognition, another programming layer of some sort would have to be added somehow to give this functionality, and that adds another level of complexity to the project that may be unwelcome. Kari actually has several other disadvantages, as well, such as not being free (I think the last time I looked, the program was like $24.00), you’re only limited to the voices that come with it, and there’s no direct way (and probably no way at all) to interact with the robot itself, to do things like use voice commands to make the robot perform certain tasks (which, I’m sure, is on your ToDo list… It would be on mine! ). At least with ChatScript, you can program in some commands to pass to the system that can do things like run a batch file that can contain all sorts of interesting things for the robot to do.

If you are going to build an actual robot, you may want to take a look at the “Robot Operating System”.
http://en.wikipedia.org/wiki/Robot_Operating_System

Since there is a javascript version of Processing, you might want to explore RiveScript.js
http://www.chatbots.org/ai_zone/viewthread/992/

My robot should be able to interact with children, so based on that alone, Kari is out.  I will be spending time learning ChatScript. Thank you for the link to the ROS and RiveScript.js. Both seem to be useful for my robot idea.