Digital assessment of speech in Huntington disease

Speech data have the potential to be a valuable digital measure of HD progression, and can also enable remote, frequent disease assessment in prodromal HD and HD. Clinical status and disease severity were predicted from extracted speech features using random forest machine learning models. Speech measurements could be leveraged as sensitive marker of clinical onset and disease progression in future clinical trials.