We envision an objective metric that quantifies the deviation of speech articulation patterns between individuals with craniofacial deformities and healthy individuals. Furthermore, while symmetry metrics can be established for still images, we hypothesize that understanding facial symmetry while speaking can provide better insights into the results of the medical procedure. 500 patients and 500 healthy subjects will be recruited in this study. Patients will read paragraphs or sentences while being recorded with regular cameras capturing the frontal view of their face. Video and audio will be recorded. The project will then use machine-learning techniques to align and compare the speech recordings with examples collected from controlled individuals.
The aim of this project is to develop an objective measurement of the speech production abilities of patients with craniofacial differences. The study will be conducted in partnership with the university of Texas at Dallas (uTD). Facial movement is a crucial aspect of speech articulation, and any changes to the facial muscles as a result of a medical procedure can greatly impact a patient's ability to effectively communicate. our goal is to investigate machine learning as a tool to quantify the differences in speech articulation patterns between individuals with and without craniofacial differences. additionally, while traditional symmetry measurements may be adequate for evaluating still images, we believe that assessing facial symmetry changes during speech will offer a more comprehensive understanding of the effects of medical procedures on speech articulation. We hope to understand these differences more thoroughly and objectively by using machine learning and patient recordings.