Natural Language Generation Essay -- Computer Science

Natural Language generationAbstract Natural address generation is a relatively rude(a) field in computer science. The two main questions to be figure out ar what to say and how to say it. What to say includes answering questions non specifically asked by the user, and remembering previous statements do by the user and by the program itself. How to say it involves construction of logical fates, use of pronouns, and sentence fragments. When diction synthesis is used, it also involves producing intelligible word rhythm, emphasis, and pauses. A new, exciting extension phone of speech synthesis is the inclusion of emotion in speech. UserELIZA workforce are all alikeIn what way?Theyre evermore bugging us about close tothing or otherCan you think of a specific example?Well, my boyfriend made me come here.Your boyfriend made you come here.He says Im depressed very much of the time.I am sorry to hear you are depressed.Its true, I am unhappy.Do you think coming here will help you not be unhappy?I need some help that much seems certain.What would it mean to you if you got some help?Perhaps I could get on with my mother.Tell me more about your family.My mother takes help of me.Who else in your family takes care of you?My father.Your father.You are like my father in some ways.What resemblance do you see? Natural language generation is the generation of printed text as well as synthesized speech. The transcript above is from a system that generates psychoanalytical responses to user input. This system is called ELIZA, and was developed at MIT in the 1960s. The program illustrates some of the issues of natur... ...gories such as pitch, voice speed, loudness, and pause length. She wrote a program that outturn these changes based on the emotion that it was asked to portray. Sample output from this program potentiometer be found online. When emotion can be output on a regular basis in speech synthesis, a new breed of natural language applications will b eginNatural language generation is a immature and exciting subject. There is still much research to be make to find the best way of representing a grammar, and providing a program open(a) of remembering previous statements, interpreting user input, speaking in sentences and sentence fragments, and providing a natural conversation. Even newer is the possibility of adding emotional flavor to synthesized speech. The windows that are opened by these studies present such a vast decorate of new applications that it is impossible to imagine them all.