Voice as a Biomarker of Health

Voice, a biomarker of health – The WHY

The human voice is often referred to as a unique print for each individual and contains biomarkers that have been linked to various diseases ranging from Parkinson’s disease to dementia, mood disorders and cancers.

Voice contains complex acoustic markers that depend on the coordination between respiration, phonation, articulation, and prosody. Recent advances in acoustic analysis technology, in particular those linked to machine learning, have shed new insights into the detection of diseases.

As a biomarker, voice is unique, cost-effective, easy and safe to collect in low resource settings. Moreover, the human voice not only contains speech, but also other acoustic biomarkers such as respiratory sounds, and cough.

A growing number of AI start-ups are using voice and other acoustic data, such as cough sounds, to screen for conditions such as burnout, vocal pathologies, and COVID-19, most recently. Although the preliminary results are promising, many limitations to Voice AI research remain.

Presently, most available voice databases are of small size and questionable acoustic quality, lack data labelling for more than one condition, and often represent a single homogeneous population. Voice is considered a biometric identifier subject to HIPAA regulation, limiting multi-institutional collaborations due to ethical considerations – ultimately hampering the creation of accessible, robust, and diverse voice datasets.

Offering new insights into disease detection

Our Aims

For voice to emerge as one of the biomarkers of health, there is a pressing need for a large, high quality, multi-institutional and diverse voice database linked to other health biomarkers from various data of different modality (demographics, imaging, genomics, risk factors, etc.) to fuel voice AI research and answer tangible clinical questions. Such endeavor is only achievable through multi-institutional collaborations between voice experts and AI engineers, supported by bioethicists and social scientists, to ensure the creation of ethically sourced voice databases representing our populations.

Our group aims to integrate the use of voice as biomarker of health in clinical care by generating a substantial multi-institutional, ethically sourced, and diverse voice database linked to multimodal health biomarkers to fuel voice AI research and build predictive models to assist in screening, diagnosis, and treatment of a broad range of diseases. Data collection will be made possible by a software through a smartphone application linked to electronic health records (EHR) and other health biomarkers such as radiomics, and genomics, and supported by federated learning technology to protect data privacy.

Fulfilling the need for a robust voice database