Smith–Magenis syndrome (SMS) is a rare genetic disease characterized by intellectual disability, serious behavior disorders, neurodevelopment delay, and speech and language disorders. An acoustic and biomechanical analysis of the voice of SMS young adults was carried out due to (a) the close relationship between the laryngeal biomechanics and the clinical and emotional state of a person; (b) the fact that no research on the voice in this syndrome has been conducted previously. The vocal timbre of most people diagnosed with SMS does not seem to be according to the complexion of diagnosed individuals, nor to their gender and age, so it could be interesting to attend the analysis of phonation of people with a rare genetic syndrome such as SMS. We used BioMetPhon, a specific piece of software to analyze the glottal source and biomechanics of vocals folds. Nineteen features related to dysphonia, physiology, and biomechanics of the vocal folds were considered. The adult phonation of 9 individuals with SMS was analyzed and compared to 100 normative male and female adult voices. Results showed that the phonation of the SMS group significantly deviates from the adult normophonic profile in more than one of the 19 features examined, such as stiffness of the thyroarytenoid muscle and dynamic mass of the vocal fold cover, among others.

Smith–Magenis syndrome (SMS) is a rare, underdiagnosed condition due to limited public awareness of genetic testing and a lengthy diagnostic process. Voice analysis can be a noninvasive tool for monitoring and detecting SMS. In this paper, the cepstral peak prominence and mel-frequency cepstral coefficients are used as disease monitoring and detection metrics. In addition, an efficient neural network, incorporating synthetic data processes, was used to detect SMS in a cohort of individuals with the disease. Three study cases were conducted with a set of 19 SMS patients and 292 controls. The three study cases employed various oversampling and undersampling techniques, including SMOTE, random oversampling, NearMiss, random undersampling, and 16 additional methods, resulting in balanced accuracies ranging from 69% to 92%. This is the first study using a neural network model to focus on a rare genetic syndrome using phonation analysis data. By using synthetic data (oversampling and undersampling) and a CNN, it was possible to detect SMS with high levels of accuracy. Voice analysis and deep learning techniques have proven to be a useful and noninvasive method. This is a finding that may help in the complex identification of this syndrome as well as other rare diseases.

The study of the voice is a very valuable tool in the field of rare diseases. That is because the biomechanical and acoustic analysis of phonation allows to detect specific features of a particular genetic and rare syndrome. In the case of Williams syndrome (WS), it was proved that some clinical features connected to this syndrome can be observed by means of voice analysis. In order to carry out this study, 60 recordings of 12 children suffering from WS were assessed and compared to 483 normative phonations of 97 children with typical development. The methodology used to assess the voices was BioMetroPhon (Gómez et al., 2013). The results may aim to establish a narrow connection between WS phonation and its typical genotype and, in consequence, the results of this study may highlight the interdisciplinary nature of the assessment of the voice.