Looking variation and change in the mouth: developing the sociolinguistic potential of Ultrasound Tongue Imaging

Lead Research Organisation: Queen Margaret University Edinburgh
Department Name: Speech and Hearing Sciences


Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
Description Medical ultrasound scanners can be used to give an entirely new window onto speech and language if they are used to provide a real time moving image of the tongue surface, synchronised to the sounds themselves. Ultrasound Tongue Imaging ("UTI") requires a probe to be held under the chin while the speaker is talking, images being captured to computer for analysis. The value for the scientific study of speech and language is that we can find out more about the underlying activity of the speaker than is revealed by a recording of the voice. Articulation is a complex process where different articulators act in consort to generate the acoustic output, and UTI lets researchers consider the movement and shape of the tongue surface independently from other articulators, to work out the changes on the voice of different co-ordination strategies.

1.We developed methods to use ultrasound tongue imaging for the first time to provide direct evidence of speech articulation in socially varied accents (of Scottish English).
2. We showed that the recording of the physical motion of articulators during speech was no more invasive or likely to cause changes to speech production that the presence of just audio recording equipment.
3. We proved that a social difference exists in subtle aspects of the production of Scottish English consonant /r/ at the ends of words. These differences in tongue shape are associated with social class differences in Scotland, disproving claims in the literature that the differences between these shapes are too subtle to be used linguistically.
4. We proved that there are cases of weak acoustic realisation of a word-final or pre-consonantal /r/ consonant that nevertheless have strong articulatory reflexes.

UTI confirms its value as a research tool in revealing that a consonant can have "covert" articulatory differences between speaker communities, and that a speaker may makes the effort to articulate the consonant, but in such a way that it is not heard. The implications for psycholinguistic reserach on speech perception and phonological/phonetic research on the nature of targets in speech production will be explored in future work.
Exploitation Route The work is proving influential in sociolinguistics and phonetics, and in speech and language therapy.
Sectors Other

URL http://www.esrc.ac.uk/my-esrc/grants/RES-000-22-2032/outputs/Read/879bf087-fa7e-4556-8d99-85b18b6f1e76