Research

While my current academic LfbA-contract does not allocate any time towards research, I do find some freetime from time-to-time in order to expand my horizon and to titillate my mind. Since my arrival in the department in 2008, English phonetics and phonology have always interested me.

For the past couple of semesters, however, my interest has honed in on the pronunciation of Middle English (ME), in particular in Chaucer's The Canterbury Tales. While challenging, I enjoy phonemically transcribing parts of this ME text, so that others may read it aloud in its original pronunciation (or my interpretation of the original pronunciation). Let's take the first few lines of "The Prologue" as an example:

In addition to creating a phonemic transcription, I am also interested in studying how the pronunciation of English has changed over the centuries. For instance, I want to discover unfamiliar sounds. A case in point is /x/ in druːxt in the above snippet . While this sound can still be heard in some varieties of Scottish English and Irish English, it is mostly absent from the so-called standard varieties of English. I also want to discover how current familiar sounds followed a very different pattern in ME. The voicing of the two th's (i.e. /ð/ and /θ/) examplifies this point. The determiner that, for instance, had a voiceless th-sound in ME. And lastly, as my doctoral thesis dealt in part with the articulation of the 18th letter of the English alphablet, I am curious as to the general articulatory development of r in English. In ME, it is often trilled. In Shakespeare's day, a major change occurs as it "loses" it trill and becomes a central approximant. That's right, Shakespeare also articulated his r's. So, when did standard British English become non-rhotic? That is for you to discover!

Don't think my research interest in earlier forms of English keeps me hundreds of years in the past! Such a judgement would be a fallacy. I am also currently exploring how artificial intelligent (AI) can be implemented into the English-language classroom, specifically with addressing pronunciation. While the advantages of AI with the writing process has received a great deal of attention, pronunciation training has once again become the Cinderella in the room. It is for this reason, I want to explore the benefits and pitfalls of AI with this particular language skill.

Image created by ChatGPT

From lesson planning to engaging exercises with the IPA-symbols, some useful practical applications for AI in the pronunciation classroom are being sought. In one of my M.Ed. classes, students and I have experimented with ChatGPT and have tried to find uses of it for pronunciation training. In one instance, we created a fictious storyline with the addition of two unlikely characters. On the left-hand side is a cartoon merging two different protagonists from two separate stories (i.e. Little Red Riding Hood and He-Man). This comic scene uses the IPA-symbols to tell a story. Not only were visual cartoons created but also short written texts with various storylines.

As learning these symbols takes lots of practice, finding amusing and educational exercises can help to pique interest as well as improve students' command of the phonetic symbols. The end result is to assist students in improving their own spoken English. In this case with the cartoon on the left-hand side, American English is the focus.

When it comes to assessing spoken language, AI-tools tend to be less beneficial and more of a pitfall. While the results in my M.Ed. classroom experiment may have been less than satisfactory, AI is constantly enhancing and improving itself. In any case, it is also important to be aware of AI's limitations.

GASP 2025 in Leipzig

Exploring the Benefits and Pitfalls of AI’s Role in an English-language Pronunciation Classroom

Technology surrounds us in our everyday life like never before. It has never been easier for instructors to incorporate digital tools into the foreign language learning environment than today. In this workshop, we will focus on how artificial intelligence—specifically ChatGPT—can be used to support pronunciation training in the English-language classroom. From lesson planning to practical exercises, both teachers and students can benefit from its integration.

However, it is important to recognize the limitations of this technology. While an AI-tool can offer some convenience, it is less than optimal, for instance, with accurate, practical feedback on oral production. Instead, feedback from an AI-tool can often be inconsistent, raising questions about its overall effectiveness in supporting the area of pronunciation development.

This workshop will explore three key areas:

1. Lesson Planning: Can an AI-tool help save time when preparing for a pronunciation lesson, and does AI take the human element into account?


2. Exercise Conceptualization: What types of pronunciation activities can an AI-tool generate, and how reliable are these instant outputs?


3. Oral-Production Assessment: Can an AI-tool effectively evaluate a student’s spoken progress, and where does it fall short?

In order to provide us with samples for further investigation, student contributions from my own integrated language course will be made available. Phonemic transcriptions with the IPA-symbols will also be included here.

 
AI is just the latest in a long line of technological innovations, evolving at an astonishing pace. The next big leap is just around the corner, so we must stay on top of the latest developments.

AI is just the latest in a long line of technological innovations, evolving at an astonishing pace. The next big leap is just around the corner, so we must stay on top of the latest developments. Let's keep researching ...