Summary: An international consortium has established the first consensus-based framework and standardized taxonomy for vocal biomarkers.
The multi-stage Delphi effort brought together 24 global experts from Europe and North America via the eVoiceNet and NIH Bridge2AI-Voice consortia. The framework resolves persistent terminology fragmentation, such as conflating voice, speech, and vocal measures, by establishing clear distinctions between unvalidated vocal measures and validated vocal biomarkers.
This structured model provides a shared scientific vocabulary across physiological, cognitive, acoustic, and computational domains to accelerate regulatory approval, clinical validation, and digital health deployment for conditions including Parkinson’s, Alzheimer’s, depression, heart failure, and type 2 diabetes.
Key Facts
- First Consensus Framework: Establishes the inaugural standardized terminology and classification model for health technology applications using human voice, speech, and respiratory acoustic signals.
- Consensus Process: Developed through a multi-stage Delphi consensus methodology (2024–2025) uniting 24 international experts across clinical medicine, speech-language pathology, acoustic engineering, data science, and regulatory affairs.
- Conceptual Distinction: Formally differentiates raw or processing-derived “vocal measures” (e.g., fundamental frequency, jitter, pause duration) from clinically validated “vocal biomarkers” (vocal features demonstrated to correlate reliably with a specific health condition or physiological state).
- Multimodal Diagnostic Breadth: Addresses disease detection spanning neurological (Parkinson’s, Alzheimer’s), psychiatric (major depressive disorder), cardiovascular (heart failure), and metabolic domains (type 2 diabetes).
- Regulatory and Clinical Translation: Designed to streamline approval pathways with bodies like the FDA and EMA by providing clear ontologies, classification standards, and logistics frameworks.
Source: USF
A person’s voice can express far more than just what they are saying. Researchers are increasingly discovering that subtle changes in speech, breathing and vocal quality provide valuable clues about a wide range of health conditions, from Parkinson’s and Alzheimer’s to depression, heart failure and type 2 diabetes.
These voice-derived indicators, known as vocal biomarkers, are expected to soon be used for disease diagnosis and monitoring.
To help unlock this potential, researchers and clinicians from the Department of Precision Health (DoPH) at the Luxembourg Institute of Health (LIH) and the University of South Florida Morsani College of Medicine have led a new international effort to establish the first consensus-based framework and definitions for vocal biomarkers.
Published in the journal Digital Biomarkers as part of the VOCAL (Vocal Biomarker Guidelines for Ontology, Classification, Application and Logistics) initiative, the study brings together 24 international experts from Europe and North America to address a key challenge facing voice-based health technologies: the lack of a common scientific language.
As research on vocal biomarkers becomes more popular, the field’s rapid growth has led to inconsistent terminology, with concepts such as “voice biomarkers,” “speech biomarkers” and “vocal biomarkers” often used interchangeably, despite referring to different physiological and cognitive processes.
To address this challenge, eVoiceNet — a European network coordinated by the Luxembourg Institute of Health, and Bridge2AI-Voice — a North American consortium funded by the NIH and co-led by USF researchers — conducted a rigorous multi-stage consensus process between 2024 and 2025.
The result is a structured framework that clearly distinguishes between vocal measures and validated vocal biomarkers and introduces a hierarchical model spanning the different domains involved in voice and speech production.
The framework provides a scientifically grounded vocabulary designed to improve collaboration between clinicians, speech and language specialists, engineers, data scientists, regulators and industry stakeholders. It also aims to support the future development of standards, validation pathways and regulatory guidance for voice-based health technologies.
“Voice has enormous potential as a source of health information, but the field cannot progress efficiently without a common language,” said Dr. Guy Fagherazzi, head of the Department of Precision Health at the LIH and chair of eVoiceNet.
“By defining what we mean when we talk about voice-based health measures, we are creating the foundations for more robust research, greater transparency and, ultimately, clinically useful technologies that can benefit patients.”
Dr. Yael Bensoussan, associate professor of Otolaryngology at the USF Health Morsani College of Medicine and co-head of the Bridge2AI-Voice consortium, said the framework reflects both the promise and complexity of vocal biomarker research.
“One of the unique strengths of vocal biomarkers is that they capture information from multiple physiological and cognitive systems simultaneously,” Bensoussan said. “However, this complexity is also what has made the field difficult to define. This work provides a structure that allows researchers to speak the same scientific language while preserving the richness of the signal.”
The publication marks the first phase of the broader VOCAL initiative, which aims to establish international guidelines and standards for vocal biomarker research and implementation. The researchers hope that a shared vocabulary will help accelerate the translation of voice-based technologies from the lab to the clinic.
Key Questions Answered:
A: A vocal measure is any quantifiable acoustic, linguistic, or respiratory parameter derived from voice or speech recordings (such as pitch perturbation or pause length). A vocal biomarker is a specific vocal measure or set of measures that has undergone formal clinical validation to reliably indicate a specific physiological state, biological process, or disease diagnosis.
A: Rapid field growth led to interchangeable and imprecise uses of terms like “voice,” “speech,” and “vocal” biomarkers. Because voice production involves complex overlapping physiological systems (laryngeal, respiratory, neurological, and cognitive), a unified taxonomy is essential for clinicians, engineers, data scientists, and regulators to collaborate, compare study results, and achieve clinical approval.
A: Research shows vocal biomarkers can track a diverse array of conditions, including neurodegenerative disorders (Parkinson’s, Alzheimer’s), mental health conditions (depression, anxiety), cardiovascular illnesses (heart failure, pulmonary congestion), and metabolic conditions (type 2 diabetes).
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- Journal paper reviewed in full.
- Additional context added by our staff.
About this AI research news
Author: Cody Hawley
Source: USF
Contact: Cody Hawley – USF
Image: The image is credited to Neuroscience News
Original Research: Open access.
“Consensus-Based Definitions for Vocal Biomarkers: The International VOCAL Initiative” by Mégane Pizzimenti, Ayush Kalia, Jamie A. Toghranegar, Mohamed Ebraheem, Nicholas Cummins, Satrajit S. Ghosh, James T. Anibal, Rhoda Au, Arian Azarang, Ruth H. Bahr, Sybille Barvaux, Steven D. Bedrick, Hugo Botha, Oita C. Coleman, Abir Elbeji, Lampros C. Kourtis, Anaïs Rameau, Jaskanwal Deep Singh Sara, Stephanie W. Watts, Daria Hemmerling, Jiri Mekyska, Marisha L. Speights, Jean-Christophe Bélisle-Pipon, Yael E. Bensoussan, Guy Fagherazzi. Digital Biomarkers
DOI:10.1159/000553327
Abstract
Consensus-Based Definitions for Vocal Biomarkers: The International VOCAL Initiative
Introduction: Voice-based health technologies are growing rapidly, but they lack standardized terminology, which hinders interdisciplinary collaboration, research quality, and clinical translation. The objective of this work is to develop universally accepted definitions in the rapidly evolving field of vocal biomarkers, as part of the VOCAL (Vocal Biomarker Guidelines for Ontology, Classification, Application, and Logistics) initiative, a structured, international consensus-based framework that aims to provide standards and guidelines.
Methods: VOCAL is a rigorous, international, multi-stage consensus-building study conducted in 2024–2025. It is a multi-institutional collaboration between representatives from the Bridge2AI-Voice Consortium (North America) and the eVoiceNet Network (European Union), involving a group of 24 international experts in medicine, clinical research, speech and language, audio signal processing, statistics, methodology, regulation, and ethics. VOCAL’s iterative process involved five rounds of review and feedback, and an in-person workshop at the 2025 Bridge2AI Voice Symposium.
Results: Consensus-based definitions for vocal biomarkers were developed, spanning from broad concepts to domain-specific measures. A hierarchical continuum model of vocal biomarkers was established. We first distinguished between the concepts of vocal measures and vocal biomarkers. We then defined terms from broad, overarching concepts (Level 0: Biomarker, Digital Biomarker, Vocal Biomarker) to more specific physiological and cognitive domains (Level 1: Cardio-Respiratory Acoustic; Level 2: Voice; Level 3: Speech/Articulatory; Level 4: Cognitive/Language, including linguistic and paralinguistic subtypes).
Conclusion: This work provides a shared vocabulary that is essential for fostering communication through interdisciplinary collaboration, improving the quality and efficiency of research and development, and ensuring the ethical, reliable, and scalable deployment of future voice-based health technologies. It lays foundational groundwork for upcoming guidelines and standards, which are crucial for advancing the field of vocal biomarkers into widespread clinical utility.
