Replacing the tutor: multimedia enhancement of articulatory phonetics training
Ian R. Smith
York University
<iansmith@yorku.ca>
Introduction
Funding cuts to tertiary education have affected us all in recent history, although some parts of the world and some disciplines have suffered more than others. Within linguistics programmes, phonetics is often under pressure because it is so expensive to teach in the traditional way, with small classes that provide a good amount of individual feedback to students. The need for effective self-study materials is thus more urgent than ever.
One of the earliest such aids was the audio tape, such as The Sounds of the IPA (produced at University College London by John Wells and Jill House; http://www.phon.ucl.ac.uk/home/wells/cassette.htm). While very helpful as an initial orientation, audio tapes suffer from three limitations that make them less useful for intense practice.
Video tapes were an advance over audio tapes in that they eliminated the first of these drawbacks, but before they came into widespread use in phonetics courses, they were overtaken by digital video on personal computers. Small, grainy and jumpy in the early 90's, computer video has seen continual improvement in size, resolution and frame rate, so that it is now useful for demonstrating articulations. In 1994 I undertook the development of a CD-based multimedia resource, subsequently dubbed Higgins the phonetics tutor, which would comprise such digital video demonstrations together with a pedagogically useful interface.
Description of the software
Finally completed at the end of 1998, Higgins presents 226 consonants, 70 vowels, and various prosodic phenomena. Sounds are organised into approximately 400 small groups based on similarity, with each sound appearing in several overlapping groups. The initial menu screen (fig. 1) offers students a choice among phenomena to work with (consonant places and manners, consonant phonation etc.) Each of the choices under Consonants and Vowels leads to a second screen where a specific group of sounds can be selected. Fig. 2 shows the consonant places and manners screen; the coloured buttons around the edge of the chart can be clicked to select the corresponding row or column of sounds. When such a group is selected, a viewing screen appears in which the sounds of the group are displayed. Fig. 3 shows the viewing screen displaying the sounds of the sixth row of the chart in Fig. 2.
Figure 1. The main menu.
Figure 2. The consonant places and manners chart.
Figure 3. The viewing screen.
Now students may view a video demonstration of any of the sounds by clicking on the corresponding symbol. (In Fig. 3 the voiceless pharyngeal fricative is being played.) Students may choose to view the sound in any combination of the environments #_V, V_# and V_V. The fact that the sounds of the group may be played in any order allows them to be compared easily. Students may also record their own pronunciations and compare them with the model. When familiar with a group of sounds, students may activate a quiz, in which the sounds of the group are played in random order for them to identify. Finally, each sound is associated with a "similar sounds group", consisting of sounds which share various articulatory or acoustic parameters with the sound in question; students navigate to this group by clicking on the "similar sounds" button.
Impact on course structure
From a purely pedagogical point of view, Higgins (like other self-study materials) would ideally be used as an adjunct to the traditional format of course delivery, to provide even more effective training. What we have done at York, however, is to use it to change our course to a less labour-intensive and hence less costly format. Indeed university funding to fully develop Higgins was contingent on the expectation that it would result in such savings.
What Higgins does is to allow individual study in the computer laboratory to be substituted for some of the hours of intense interaction with a human instructor. Prior to the introduction of Higgins, phonetics was taught at York in groups of 25 students (maximum enrolment) meeting three hours a week. With Higgins, the course is structured around a two-hour lecture, a one-hour tutorial and a drop-in computer laboratory. Tutorials have a maximum enrolment of 16, while lecture size is limited only by the size of the room (which must be equipped with closed-circuit video facilities for demonstration). In terms of personnel, savings are realised once student numbers exceed 25 (in the old format, 6 instructor-hours would be required for 26-50 students; under the new régime 4 hours are needed for 16-32 students, 5 for 33-48). The expense of providing the computer laboratory, which should also be included in calculating the overall cost of course delivery, is ignored here, as most universities seem happy to treat such laboratories as part of essential infrastructure rather than as annual debits to the pedagogical budget.
Pedagogical impact
At York University, Higgins has been integrated into in a second-year, one-semester course in articulatory phonetics for linguistics majors. Currently, class size is approximately 50-70. Higgins was used in the 1998-99 academic year and, in a preliminary form, in the previous year.
As noted above, Higgins, has also allowed us to restructure the course to provide for fewer instructor hours (while giving students the same number of class hours). This change muddies the waters for any quantitative assessment of the educational value of Higgins. Other variables difficult to control include personnel and test format. Impressionistically, however, the educational outcomes do not appear to be significantly altered by the new course format.
Most importantly, student reaction solicited as part of an anonymous assessment of various aspects of the course this past term was overwhelmingly positive. In addition to noting the general effectiveness of Higgins, some students commented that they feel intimidated by a human tutor and are more relaxed working with a computer; others noted that Higgins allows them to work at their own pace and to concentrate on their own particular needs.
Conclusion
Although the primary motivation for developing Higgins was pedagogical, it has been an effective, student-friendly substitute for some of the hours previously spent with a human instructor.
Higgins is particularly good for reinforcing knowledge of symbols and sounds and for developing skill at discriminating auditorily and visually between different sounds. It also provides limited feedback through the self-record function.
While Higgins overcomes the first two limitations of audio tape, it does not completely replace the tutor in providing feedback. Students vary in their ability to make use of the playback of self-recordings – some are better than others at hearing whether their own productions match the model or not. We still need a human to say "move your tongue forward" or "your velum is drooping".