format_quote Originally Posted by
Sum-Muslim-Gal
salam
ok am an A2 student studying English lang...
I based my course work on Artificial Intelligence[AI]....
It involves robots talking ...even one that actually sing to lol...
anyways I don’t have big hypothesis yet although I’ve though of one ' do machines understand every context of conversation.??...then I will look at the frameworks from linguistic perspective, maybe psychologist n sociologist perspective too...
ok ive been talking to my teacher she's thinking it be good if I have a second human into my investigation so that I can divide the bots I talk to 3 & the second human talk to 2 bots. well something like that…
well your help comes in...i need a second human!!....
my friends are busy with exam revision or their own stress on c/work so i don’t want to bother them...I don’t want to bother u guys either lol….but if anyone is willing to help me out here I’ll be so great full…in your own good time…that’s if your not busy with anything…please!!!
Sorry am not elaborating in depth …am kind of in rush!!..
inshaAllah please post if u would like to…:D:D:D
& I will get back to you soon as…
JazakaAllah
Wasalam
Virtual Hindi Chat
Deepti, the Talking Computer – By Rahul Jindal
Chandigarh students have developed a program which lets computers talk in Hindi. It has enormous potential in India, writes Rahul Jindal, who developed the program with his associates.
Hello India. Artificial Intelligence, meet India’s illiterate peasant. Four of us, originally from Chandigarh’s Punjab Engineering College, are now building what we believe could be a vital bridge that unites India’s proud, state-of-the-art computer expertise with its impoverished masses.
Imagine a scene at a busy railway station in New Delhi. A semiliterate passenger wants to know what time the train to Jalandhar will leave. He goes to the information booth, finds the monitor which has the information. Maybe he chooses the distinctive icon that represents the his train. Or he chooses the train number. He touches it, and a voice tells him what time the next train shall leave. It can also guide him to more information about reservations and availability.
Sounds like a sci-fi movie? Well, it isn’t, thanks to Deepti. Touchscreen technology is now commonplace in the West, and Deepti is a program that we have developed which is an interactive computer chat program that can speak in Hindi.
The three-member team includes me (Rahul Jindal), Ritvik Sahajpal and Rohit Kumar working with Rishi Bihari. We have come up with a chat robot — a chat software program—— that understands and speaks Hindi and is capable of intelligent talk.
It has great potential, we believe. Deepti could be used to make government services more accessible. We could put it in government offices, where it could replace the information desk. Fully developed, we could have a system where people could communicate with it via voice and it could then provide them help, say, to get the right official forms, again through voice.
Deepti could also be used on home computers. Users could install it on their computers and have fun chatting with Deepti. As Deepti chats more and more with the user, it will acquire more knowledge about the user and become more and more intelligent.
The ability to communicate through speech is something we take for granted, but it is one of the most complex of skills, as scientists who do research in artificial intelligence have discovered. Talented computer scientists are now on a path to teach computers to do this. The ability of computers to talk comes through the use of Natural Language Processing techniques. The simplest of these is based on the pioneering work of the AI Foundation which has developed an as-yet-primitive non-standard “markup” language for artificial intelligence, called AIML. AIML is the brainchild of Dr. Richard S. Wallace, a computer scientist with Carnegie Mellon University
Scientists are trying to develop chat robots using this, and substantial success has been achieved.
Our project works best through modular programming. We wish to follow the dictum “From simple to complex” throughout the development process. We would wish to set out with a study of AIML along with planning and discussing the simplest of the AIML categories.
The development and conception of these categories will undoubtedly be a continuous process.
We will need to develop what might be called an AIML server, that understands the AIML rules and hence our AIML categories. This is actually what the “robot” is. With a weak “shell” developed, we move on to giving a simple command line interface (aka client), which would in fact be an interface to the human user. It is as if we take our server to be analogous to a http Server and the User Interface as being analogous to an http Client that sends requests to the server on behalf of the user.
Until now both the user and the bot — a machine that communicates through typed speech — communicate through Unicode (or ASCII) characters. At this stage we are working towards transforming the bot output to simulate human voice. This can be done using a Text-to-Speech system. Currently we work with user inputs in text format with the bot “speaking” out her dialogues.
There is a lot of work ahead. We would like to reiterate that categories of AIML would need to be refined to make Deepti intelligent.
We would also like to move from command line interface to a Web browser-compatible interface. For this we would need to direct all input to a software port at which our server would be listening, so that appropriate responses can be played on a multimedia-enabled system through the speakers.
As we develop our system, we wish to ultimately have some form of lip syncing, so that we have a character on screen that seems to be speaking out her responses.
Few things are certain in scientific research. The thorniest problems can be solved in unexpected flashes of inspiration; on the other hand, unforeseen hurdles can come up and stymie progress. But perseverance and hard work will get you there.
We are very, very optimistic. The basic principles are sound, and we already have a prototype, though it needs refining. So it may not be too far in the future when at a major Indian railway station, when you need to know about train schedules, you may be looking at the face of a beautiful (virtual) woman on a monitor who answers your questions with seductive grace. Now that would be an improvement over the usual scene today: A dour employee who wordlessly points a stick at some blackboard as he looks at you accusingly, as if asking about the train schedule is an only slightly less serious crime than culpable homicide.
– Rahul Jindal presently works with Infosys in India.
He helped develop Deepti, a software program
that enables computers to talk in Hindi..