Ta-ka-ti-ki Update

3 July 2013 by Kerstin Hoppenhaus, posted in Linguistics

In my first week at the institute, I served as a test subject in the phonetics lab, where Leonardo Lancia and Sven Grawunder are developing a new statistical tool for analyzing ultrasound images.

These are some of the images that came out of these trials.

To me, they look rather cryptic, but he scientists were quite excited about their results.

The test showed that the pictures are clear enough and the various outside parameters are sufficiently controlled to allow for a reliable statistical analysis.

Which is great. Because with this, they will now be able to tackle more complicated structures. First up is the larynx. It plays an important role in vocalization, but since it sits deep down in the throat it has been notoriously difficult to study.

This may all seem rather academic, but to understand the physiological determinants of language is actually quite useful. For example, when you study the structure of languages, it would be very helpful to be able to separate features that have developed out of communicative needs or historical conditions from those that are simply caused by physiological constraints. Or, if you study ancient humans, understanding the exact role of physiology in language production would help a lot to better assess the ancients’ capabilities for speech.

Despite these useful applications, much of this hasn’t been done yet, mostly for technical reasons. You have to look at the process when it happens. It’s all deep inside heads and throats. And it all happens very fast. Ultrasound, of course, has been around for a while, but automatic image recognition is still difficult and there is no way even remotely efficient enough to analyse this kind of data manually.

Which is, of course, why Leonardo and Sven are so pleased with their new tool. For them, it opens a new path into uncharted linguistic territories. Over the next months, they will be testing all sorts of speakers in Leipzig (one advantage of an internationally cast research institute), but ultimately, they are planning to take their setup on the road, to more remote corners of the planet, to the small and precious languages residing there.

Leave a Reply


× two = 8