Ultrax Speech Sound Disorders
A dataset of ultrasound and audio recordings from children with speech sound disorders
The UXSSD dataset contains 8 speakers (2 female and 6 male), aged 5-10 years.
The table below give further details for each speaker. Ages were taken in the first Assessment session and are indicated in years (AGE-Y) and months (AGE-M).
|BL||Baseline session before therapy (1-2 sessions)|
|Mid||Mid-point session, halfway through therapy|
|Post||Post-therapy session, immediately after therapy ended|
|Maint||Maintenance session, some time after therapy ended|
Core data types
|ult||raw ultrasound data|
|txt||prompt text with date/time of utterance recording|
All labels are provided in Praat's TextGrid format.
|slt-labels||manual annotation from SLT, when available. See  for details|
|speaker-labels||speaker diarization identifying therapist (SLT) and child (CHILD) speech|
|word-labels||automatic word-level alignment|
|phone-labels||automatic phone-level alignment|
Individual recordings are indexed for each session according to their recording times. See the prompt text file for recording date/time.
Each file ID also includes a prompt type identifier. See Data for details.
Speaker 05M was subjected to two rounds of therapy, with corresponding Assessment sessions. These are identified as *_round2 in the speaker directory. Therapy sessions for this speaker are indexed chronologically.
 Eshky, A., Ribeiro, M. S., Cleland, J., Richmond, K., Roxburgh, Z., Scobbie, J., & Wrench, A. (2018) Ultrasuite: A repository of ultrasound and acoustic data from child speech therapy sessions. Proceedings of INTERSPEECH. Hyderabad, India.
 Cleland, J., Scobbie, J. M., & Wrench, A. A. (2015). Using ultrasound visual biofeedback to treat persistent primary speech sound disorders. Clinical linguistics & phonetics, 29(8-10), 575-597.