The King's Voice and Speech Processing for Health Group conducts pioneering research to advance the integration of speech technologies and biomarkers into clinical research and practice.

Guided by a responsible Artificial Intelligence (AI) framework, our work combines cutting-edge speech processing techniques, patient and public involvement and engagement (PPI-E), advanced statistical analytics and machine learning.

This multidisciplinary approach aims to harness the transformative potential of speech technologies, creating reliable tools that drive innovation in health research and clinical applications.

Our research focuses on three core themes:

• We conduct exploratory studies to investigate voice and language biomarkers for assessing symptoms and health states in mental and neurological conditions. We collect and analyse speech samples in the lab and clinic, and remotely with digital mobile health tools.
• We undertake methodological research to examine factors within speech recording and analysis pipelines, evaluating their impact on the reliability of health assessments.
• We test and validate AI-driven speech technologies to assess their reliability and effectiveness across diverse patient populations and operating conditions.

Embedded within the Institute of Psychiatry, Psychology & Neuroscience, we collaborate extensively across international and multidisciplinary projects, including RADAR-CNS, ART-CARMA, AIMS-2-TRIALS and STORY. These initiatives span diverse conditions, including depression, multiple sclerosis, attention-deficit/hyperactivity disorder, autism and eating disorders. We welcome new collaborations to advance our research and drive new groundbreaking discoveries.