Hi Markamas & Maxhirez,I saw
Hi Markamas & Maxhirez,
I saw Markamas’ post and I couldn’t resist … I just Borg’d this (https://code.google.com/p/chatter-bot-api/) into MRL - So now it has a cleverbot service. The API was done very well, and it has bindings to .net, Python, & php - it was written in Java which is nice.
The example the author created was a bit entertaining too… his API allows communication to chatterbot and pandobots (http://www.pandorabots.com/botmaster/en/home) so the example does a dialog between the two
For me this is what happened.
bot1: Hi
bot1: Hello! Have you a question fo me?
bot2: Hi thee. How ae you? OK, hee is a question fo you … What do people do in clock factoies?
bot1: I don’t question it, Damon.
bot2: What do people do in clock factoies? They make faces all day. Ha, ha!
bot1: Cuelty. A lot of people ae senselessly inhuman to one anothe. It’s vey disconceting.
bot2: cuelty? Biefly, cuelty means a cuel act; a delibeate infliction of pain and suffeing. That is a big genealization about people. Is it?
bot1: That made almost no sense.
bot2: Is this anything to do with people?
bot1: Unless you’e going to offe me a ed pill and a blue pill, ni.
bot2: Neve mind that! Shall I tell you a poem?
bot1: Yes please.
bot2: The joke you just told isn’t funny one bit, bot1: I did not undestand that…
bot2: Why didn’t you?
bot1: Because thee was none available. Pie is not synonymous to leftoves.
bot2: I wonde why. Soy, what did you say?
bot1: I said you’e not vey cleve.
bot2: OK, I head you!
bot1: Well I did not talk to you yesteday, and my name is Clevebot.
bot2: Why didn’t you?
bot1: Because I don’t have to know you name.
bot2: That’s coect, you don’t have to!
bot1: I know you don’t think.
bot2: As Descates said, I am, theefoe I think.
bot1: What does that mean to you?
bot2: What do you not undestand?
bot1: You.
bot2: OK. Let’s talk about me.
bot1: What do you like to do.
bot2: I like chatting with people online.
bot1: Why do you like chatting with me?
bot2: That’s just the way I am.
bot1: Too tue.
bot2: Soy, what did you say?
bot1: We don’t need no education!
So, I believe it’s a bit like chaos theory where the outcome has very high dependence on initial conditions and some random generators… as do all stoner conversations… to test this I ran it again…
bot1> Hi
bot2> Hi! Can I ask you a question?
bot1> Ok.
bot2> Have you ever seen a UFO?
bot1> No but my dad has.
bot2> Do you think UFOs exist?
bot1> No.
bot2> OK. What do you call fishing when you don’t catch any fish?
bot1> I would think that you don’t get along well with humans.
bot2> What do you call fishing when you don’t catch any fish? Drowning worms. Ha ha!
bot1> I think that you think that you do not want to share what you think deep inside.
bot2> I am always interested in your thoughts.
bot1> Oh. What do you think about Antartica?
bot2> Have you got a pain? I try to think as little as possible.
Sphinx is already in MRL - but I’ve never used Sphinx to recognize free-form utterances - usually you use a grammar file - There’s another Google STT service - but it seems kind of slow (it posts sound files to google)… fun stuff though
Nice to see you’all on the ether…