Abstract: A world consortium has established the primary consensus-based framework and standardized taxonomy for vocal biomarkers.
The multi-stage Delphi effort introduced collectively 24 international consultants from Europe and North America through the eVoiceNet and NIH Bridge2AI-Voice consortia. The framework resolves persistent terminology fragmentation, similar to conflating voice, speech, and vocal measures, by establishing clear distinctions between unvalidated vocal measures and validated vocal biomarkers.
This structured mannequin gives a shared scientific vocabulary throughout physiological, cognitive, acoustic, and computational domains to speed up regulatory approval, medical validation, and digital well being deployment for circumstances together with Parkinson’s, Alzheimer’s, despair, coronary heart failure, and kind 2 diabetes.
Key Details
- First Consensus Framework: Establishes the inaugural standardized terminology and classification mannequin for well being know-how functions utilizing human voice, speech, and respiratory acoustic indicators.
- Consensus Course of: Developed by means of a multi-stage Delphi consensus methodology (2024–2025) uniting 24 worldwide consultants throughout medical medication, speech-language pathology, acoustic engineering, information science, and regulatory affairs.
- Conceptual Distinction: Formally differentiates uncooked or processing-derived “vocal measures” (e.g., elementary frequency, jitter, pause length) from clinically validated “vocal biomarkers” (vocal options demonstrated to correlate reliably with a selected well being situation or physiological state).
- Multimodal Diagnostic Breadth: Addresses illness detection spanning neurological (Parkinson’s, Alzheimer’s), psychiatric (main depressive dysfunction), cardiovascular (coronary heart failure), and metabolic domains (kind 2 diabetes).
- Regulatory and Scientific Translation: Designed to streamline approval pathways with our bodies just like the FDA and EMA by offering clear ontologies, classification requirements, and logistics frameworks.
Supply: USF
An individual’s voice can specific excess of simply what they’re saying. Researchers are more and more discovering that delicate modifications in speech, respiratory and vocal high quality present worthwhile clues about a variety of well being circumstances, from Parkinson’s and Alzheimer’s to despair, coronary heart failure and kind 2 diabetes.
These voice-derived indicators, often known as vocal biomarkers, are anticipated to quickly be used for illness prognosis and monitoring.
To assist unlock this potential, researchers and clinicians from the Division of Precision Well being (DoPH) on the Luxembourg Institute of Well being (LIH) and the College of South Florida Morsani School of Drugs have led a brand new worldwide effort to ascertain the primary consensus-based framework and definitions for vocal biomarkers.
Printed within the journal Digital Biomarkers as a part of the VOCAL (Vocal Biomarker Pointers for Ontology, Classification, Software and Logistics) initiative, the research brings collectively 24 worldwide consultants from Europe and North America to handle a key problem going through voice-based well being applied sciences: the shortage of a typical scientific language.
As analysis on vocal biomarkers turns into extra in style, the sphere’s speedy development has led to inconsistent terminology, with ideas similar to “voice biomarkers,” “speech biomarkers” and “vocal biomarkers” typically used interchangeably, regardless of referring to completely different physiological and cognitive processes.
To deal with this problem, eVoiceNet — a European community coordinated by the Luxembourg Institute of Well being, and Bridge2AI-Voice — a North American consortium funded by the NIH and co-led by USF researchers — performed a rigorous multi-stage consensus course of between 2024 and 2025.
The result’s a structured framework that clearly distinguishes between vocal measures and validated vocal biomarkers and introduces a hierarchical mannequin spanning the completely different domains concerned in voice and speech manufacturing.
The framework gives a scientifically grounded vocabulary designed to enhance collaboration between clinicians, speech and language specialists, engineers, information scientists, regulators and trade stakeholders. It additionally goals to assist the long run growth of requirements, validation pathways and regulatory steerage for voice-based well being applied sciences.
“Voice has huge potential as a supply of well being data, however the subject can’t progress effectively with out a frequent language,” mentioned Dr. Man Fagherazzi, head of the Division of Precision Well being on the LIH and chair of eVoiceNet.
“By defining what we imply after we speak about voice-based well being measures, we’re creating the foundations for extra sturdy analysis, better transparency and, finally, clinically helpful applied sciences that may profit sufferers.”
Dr. Yael Bensoussan, affiliate professor of Otolaryngology on the USF Well being Morsani School of Drugs and co-head of the Bridge2AI-Voice consortium, mentioned the framework displays each the promise and complexity of vocal biomarker analysis.
“One of many distinctive strengths of vocal biomarkers is that they seize data from a number of physiological and cognitive methods concurrently,” Bensoussan mentioned. “Nonetheless, this complexity can be what has made the sphere tough to outline. This work gives a construction that permits researchers to talk the identical scientific language whereas preserving the richness of the sign.”
The publication marks the primary section of the broader VOCAL initiative, which goals to ascertain worldwide pointers and requirements for vocal biomarker analysis and implementation. The researchers hope {that a} shared vocabulary will assist speed up the interpretation of voice-based applied sciences 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 (similar to pitch perturbation or pause size). A vocal biomarker is a selected vocal measure or set of measures that has undergone formal medical validation to reliably point out a selected physiological state, organic course of, or illness prognosis.
A: Speedy subject development led to interchangeable and imprecise makes use of of phrases like “voice,” “speech,” and “vocal” biomarkers. As a result of voice manufacturing entails advanced overlapping physiological methods (laryngeal, respiratory, neurological, and cognitive), a unified taxonomy is crucial for clinicians, engineers, information scientists, and regulators to collaborate, examine research outcomes, and obtain medical approval.
A: Analysis reveals vocal biomarkers can monitor a various array of circumstances, together with neurodegenerative problems (Parkinson’s, Alzheimer’s), psychological well being circumstances (despair, anxiousness), cardiovascular sicknesses (coronary heart failure, pulmonary congestion), and metabolic circumstances (kind 2 diabetes).
Editorial Notes:
- This text was edited by a Neuroscience Information editor.
- Journal paper reviewed in full.
- Extra context added by our workers.
About this AI analysis information
Writer: Cody Hawley
Supply: USF
Contact: Cody Hawley – USF
Picture: The picture is credited to Neuroscience Information
Unique Analysis: Open entry.
“Consensus-Primarily based Definitions for Vocal Biomarkers: The Worldwide 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, Man Fagherazzi. Digital Biomarkers
DOI:10.1159/000553327
Summary
Consensus-Primarily based Definitions for Vocal Biomarkers: The Worldwide VOCAL Initiative
Introduction: Voice-based well being applied sciences are rising quickly, however they lack standardized terminology, which hinders interdisciplinary collaboration, analysis high quality, and medical translation. The target of this work is to develop universally accepted definitions within the quickly evolving subject of vocal biomarkers, as a part of the VOCAL (Vocal Biomarker Pointers for Ontology, Classification, Software, and Logistics) initiative, a structured, worldwide consensus-based framework that goals to supply requirements and pointers.
Strategies: VOCAL is a rigorous, worldwide, multi-stage consensus-building research performed in 2024–2025. It’s a multi-institutional collaboration between representatives from the Bridge2AI-Voice Consortium (North America) and the eVoiceNet Community (European Union), involving a bunch of 24 worldwide consultants in medication, medical analysis, speech and language, audio sign processing, statistics, methodology, regulation, and ethics. VOCAL’s iterative course of concerned 5 rounds of evaluate and suggestions, and an in-person workshop on the 2025 Bridge2AI Voice Symposium.
Outcomes: Consensus-based definitions for vocal biomarkers have been developed, spanning from broad ideas to domain-specific measures. A hierarchical continuum mannequin of vocal biomarkers was established. We first distinguished between the ideas of vocal measures and vocal biomarkers. We then outlined phrases from broad, overarching ideas (Degree 0: Biomarker, Digital Biomarker, Vocal Biomarker) to extra particular physiological and cognitive domains (Degree 1: Cardio-Respiratory Acoustic; Degree 2: Voice; Degree 3: Speech/Articulatory; Degree 4: Cognitive/Language, together with linguistic and paralinguistic subtypes).
Conclusion: This work gives a shared vocabulary that’s important for fostering communication by means of interdisciplinary collaboration, bettering the standard and effectivity of analysis and growth, and making certain the moral, dependable, and scalable deployment of future voice-based well being applied sciences. It lays foundational groundwork for upcoming pointers and requirements, that are essential for advancing the sphere of vocal biomarkers into widespread medical utility.
