Computers and Multimedia Applications in Teaching Pronunciation

In our poster presentation we intend to demonstrate the results of the research we have carried out in the field of experimental phonetics and teaching pronunciation. The ultimate question we have asked ourselves is "to what extent can computer-based methods facilitate second language acquisition in a non-native environment". Our subjects being undergraduate students at Kossuth University, we are also interested in whether or not it is too late to start acquiring a better pronunciation (covering both segmental and suprasegmental levels) well after one has reached the so-called critical age. In the course of our research we have made extensive use of the Computerised Speech Laboratory and several multimedia programs we have made for our students.

Since the prerequisite for any further improvement is the assessment of the student's current pronunciation, the first part of the poster presentation demonstrates the most typical features of a Hungarian accent. Several English speech sounds do not have their counterparts in Hungarian, therefore areas of difficulty can often be predicted. Some of these, however, remain unnoticed by Hungarians, which is due to a phenomenon called "compensatory pronunciation". When a speaker is unable to articulate a sound properly, he or she tends to make up for the required sound quality by resorting to other manoeuvres. It is typical in the case of the English [æ], which has proved to be one of the most difficult sounds for Hungarians. Here the feature of openness is often replaced by a marked modification of the fundamental frequency, which seems to be a more convenient means of sound modification.

Another interesting phenomenon is the way pitch and loudness intermingle in the overall effect of stress for Hungarians. Some students have difficulty in realizing a marked degree of pitch-change without an exaggerated degree of loudness. Wrong nucleus-placement and nucleus-placement accompanied with an extra articulatory effort therefore also contribute to a Hungarian accent.

The second part of the presentation concentrates on the improvement of such speakers. Practically all the mistakes in pronunciation can be corrected by a method whereby the learner is provided with graphical feedback on pronunciation. This method proves very effective even when repeated audio feedback cannot yield further results.

The poster touches upon a method which is still at its experimental stage: accent booting. In this learning process the speaker is deprived of his/her pronunciation and only a native speaker's model is supplied. The main objective of this method is to eliminate what we call "error-recycling", which means that speakers inevitably tend to follow and reproduce their own foreign accent, because this is what they hear most of the time.