Listening for what matters
What are vocal biomarkers and why do they matter?
Biomarkers are objective indicators that can be measured from a patient and utilised for the development of medical screening and diagnostic procedures.
Speech is a rich source of such information about an individual’s overall health and wellbeing. However, there are many factors (such as language, recording environment and variability between individuals) which make discerning health status from speech directly a complex process.
This is where our research comes in. Our team of experts is exploring whether vocal biomarkers, developed using speech science, machine learning and large datasets of medically characterised voices, could help improve detection and monitoring of neurological conditions. This research is ongoing; our technology is still in development and has not yet been clinically validated or approved for clinical use.
If you have any questions regarding our research or opportunities with the team, please get in touch with us at [email protected].
Our approach
From speech to insight.
Building on our expertise in voice pathology, we bring together three complementary areas of research.
- 01
Speech science
Understanding the characteristics of speech and how they change with neurological conditions.
- 02
Machine learning
Developing methods to identify meaningful patterns within the complexity of the human voice.
- 03
Clinical datasets
Working with large datasets of medically characterised voices to connect speech with health.
The science behind our work
Our research.
Explore publications from our team and research collaborators.
- 2026
Odyssey
Comparator loss: An ordinal contrastive loss to derive a severity score for speech-based health monitoring
View full citation
Webber, J. J., Watts, O., Wihlborg, L., Tam, J., Weaver, C., Pal, S., Chandran, S., & Valentini-Botinhao, C. (2026). Comparator loss: An ordinal contrastive loss to derive a severity score for speech-based health monitoring. Proc. Odyssey 2026, pp. 136–143. Read publication
- 2026
IEEE ICASSP
Evaluating pretrained speech embedding systems for dysarthria detection across heterogenous datasets
View full citation
Wihlborg, L., Goodall, J., Wheatley, D., Webber, J. J., Tam, J., Weaver, C., Pal, S., Chandran, S., Seth, S., Watts, O., & Valentini-Botinhao, C. (2026). Evaluating pretrained speech embedding systems for dysarthria detection across heterogenous datasets. Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2026, pp. 14952–14956. Read publication
- 2025
Interspeech
Anne Rowling Neurological Speech Corpus: Clinically annotated longitudinal dataset for developing speech biomarkers in neurodegenerative disorders
View full citation
Tam, J., Weaver, C., Watts, O., Chandran, S., Pal, S., & Rowling Speech Consortium. (2025). Anne Rowling Neurological Speech Corpus: Clinically annotated longitudinal dataset for developing speech biomarkers in neurodegenerative disorders. Proc. Interspeech 2025, pp. 5693–5697. Read publication
- 2025
BMJ Open
Digital App for Speech and Health Monitoring Study (DASH): Protocol for a prospective longitudinal case–control observational study for developing speech datasets in neurodegenerative disorders and dementia
View full citation
Tam, J., Weaver, C., Ihenacho, A., Newton, J., Virgo, B., Barrett, S., Neale, J., Perry, D., Smith, A., Chandran, S., Watts, O., Pal, S., & DASH Consortium. (2025). Digital App for Speech and Health Monitoring Study (DASH): Protocol for a prospective longitudinal case–control observational study for developing speech datasets in neurodegenerative disorders and dementia. BMJ Open, 15(12), e100222. Read publication
- 2023
npj Digital Medicine
A systematic review and narrative analysis of digital speech biomarkers in Motor Neuron Disease
View full citation
Bowden, M., Beswick, E., Tam, J., Perry, D., Smith, A., Newton, J., Chandran, S., Watts, O., & Pal, S. (2023). A systematic review and narrative analysis of digital speech biomarkers in Motor Neuron Disease. npj Digital Medicine, 6, Article 228. Read publication