Deducing the severity of psychiatric symptoms from the human voice
This work addresses the need for automated, objective symptom grading in psychiatry to aid diagnosis and treatment, but it appears incremental as it builds on existing conjectures about voice biomarkers.
The researchers tackled the problem of objectively grading psychiatric symptom severity by using acoustic data from clinician-patient interviews and non-parametric models to predict symptom ratings based on voice features, showing that different speech units capture different symptoms.
Psychiatric illnesses are often associated with multiple symptoms, whose severity must be graded for accurate diagnosis and treatment. This grading is usually done by trained clinicians based on human observations and judgments made within doctor-patient sessions. Current research provides sufficient reason to expect that the human voice may carry biomarkers or signatures of many, if not all, these symptoms. Based on this conjecture, we explore the possibility of objectively and automatically grading the symptoms of psychiatric illnesses with reference to various standard psychiatric rating scales. Using acoustic data from several clinician-patient interviews within hospital settings, we use non-parametric models to learn and predict the relations between symptom-ratings and voice. In the process, we show that different articulatory-phonetic units of speech are able to capture the effects of different symptoms differently, and use this to establish a plausible methodology that could be employed for automatically grading psychiatric symptoms for clinical purposes.