A Computer-Assisted Learning Software to Help Teaching English to Brazilians

CAI (Computer Assisted Instruction) is a learning process with the aid of computers. In the last 40 years, there was an exponential growth in the use of computers as providers of instructions. During this fast technological revolution, the CAI has become more refined and computers turned into a sophisticated, and why not say essential, instrument of aid, to the extent that they invented a new way of teaching and learning, involving more easily the student in the universe intellectual. Over the years more and more systems are incorporating CAI multimodal interfaces such as sound, image, video and interactive programs that support a wide range of applications. More recently, started up the incorporation of speech interfaces in environments called CALL (Computer-assisted Language Learning). Despite creating new possibilities for the development of learning tools, the practical impact of CALL applications in education of foreign languages has been quite modest, consequence of evaluations and interfaces without standardization, and the absence of concrete evidences proving the educational benefits of these systems [CHA 97]. But experts, as in [WAR 96], say that with the drastic and rapid invasion of computers in schools and homes, the languages teachers need, as soon as possible, revise their concepts and work in the search of mechanisms that help to overcome the obstacles set by the limitations of these speech technologies. In recent years, the performance of "personal computers" has evolved with the production of ever faster processors, a fact that encourages the implementation of speech processing algorithms in real time and economically viable. The integration of high-quality audio, high-resolution images and videos with the output of text and graphics of conventional computers has generated recent multimedia systems that provide powerful "training tools" explored on CALL. The results of these efforts are prototypes with speech interfaces for training the human pronunciation in a wide variety of languages, texts reading and foreign languages teaching in predefined contexts [EHS 98]. An important area within the speech processing is the training of pronunciation. Intelligent tutors interact and guide students in the repetition of words and phrases, or reading sentences, in order to practice both the fluency (quality of the phonetic pronunciation), as the pitch (manipulation of the parameter [pitch]) in a certain language. These prosodics characteristics are particularly important in the language learning processing, since a incorrect prosody can block the communication between the speakers. Errors in intonation can give to the listener a misconception impression of the attitude taken by the person who is speaking, as well as a sentence without fluency makes it virtually impossible for the listener understands the dialogue context. There is not a consensus among the linguists about the presence of teachers or tutors in the process. To [FRA 00], pronunciation training softwares must be employed to enhance the interaction between teacher and student, where the tutor must concentrate their efforts on speech characteristics that may affect their comprehensibility, while [ANA 03], believes in replacement of "human tutors", and suggests that the students can follow their own learning rhythm. However, linguists may attend the technicians, suggesting which speech aspects of the practitioner need to be focused and stipulating limits for the shunting lines identified in the pronunciation. A widespread style of CALL, especially in schools of languages, makes use of an high-level interface and grammar set within a context, regular situations in real life, to practice a foreign language [CCA 98]. Basically, the student must choose a response within a limited number of alternatives shown on the screen, or may be challenged to answer a question without any help from the software. These methods are called closed response and open response, respectively. The proposed application is able to merge the use of these two methods by foreign language learning (English). The student is introduced to respond orally, or even manually, the questioning by written, spoken and visual stimulations. Besides the proposed objective and subjective exercises, there is the possibility of a prior training, where students can enhancement their vocabulary by listening Text-To-Speech individualized words or phrases,