Helping children learn non-native articulations: The implications for ultrasound-based clinical intervention

An increasing number of studies are examining the effectiveness of ultrasound as a visual biofeedback device for speech production training or therapy. However, no randomised control trials exist. We compared the success of typically-developing children learning new articulations with and without ultrasound biofeedback. Thirty children aged 6-12 were randomly assigned to 2 groups: Group U were taught novel (non-English) consonants and vowels using ultrasound in addition to imitation, modelling, articulatory descriptions and feedback on performance. Group A were taught the same speech sounds, using the same methods but in the absence of ultrasound visual biofeedback. Results showed that both groups of children improved in their production of the novel sounds with the exception of the high back vowels [u,ɯ]. No advantage for Group U was found, except for the palatal stop [c].