• Description of the project
The Case for Vocal Biomarkers in Healthcare
The recent emergence of Artificial Intelligence (AI) methods and audio signal processing techniques open new perspectives on the use of voice to detect or monitor diseases. The vocal biomarker research field is booming but is facing, like other AI-driven fields, a reproducibility and generalisability crisis. To move this promising field of research forward, there is an urgent need to develop a common research framework in Europe, with standardization principles and definitions of good practices and guidelines to help it reach its full potential. A multidisciplinary approach is necessary to tackle this complex task impacting all stakeholders in healthcare virtually.
eVoiceNet aims to establish Europe as a leader in vocal biomarkers, create a collaborative and multidisciplinary network, and facilitate knowledge sharing, standardisation, and the development of privacy-aware solutions, by creating an international network of clinicians, AI experts, speech/voice processing specialists, voice pathologists, privacy/data protection experts, go-to-market specialists, venture capitalists and other providers of financial resources, patients organisations, policymakers as well as industrial partners and other stakeholders (end users, regulators, and other decision-makers), to overcome the major challenges and boost the integration of voice technologies into clinical practice.
• Backgrounds
Unlocking the Potential of Vocal Biomarkers
The human voice is a powerful medium for communication and a promising indicator of health and well-being. Easy to collect, low-cost, and non-invasive, voice data represent a valuable resource with the potential to transform healthcare. Beyond speech, other vocal-tract-related respiratory sounds such as coughing or breathing can provide additional clinically relevant information for disease characterization and monitoring.
Driven by advances in artificial intelligence, audio signal processing, and natural language processing, voice technology is rapidly expanding, with healthcare emerging as a key application area. While some applications are already in use (such as voice-assisted clinical documentation) research increasingly focuses on identifying vocal biomarkers for disease screening, diagnosis, symptom monitoring, disease progression tracking, and treatment evaluation.
Vocal biomarkers are measurable features extracted from voice signals that are associated with clinical outcomes. Changes in voice, breathing, or cough patterns can reflect a wide range of conditions, including respiratory, neurological, neurocognitive, cardiovascular, endocrine, and mental health disorders. Although vocal biomarkers are non-invasive, scalable, and hold strong potential for early detection and personalized medicine, their reliability can be affected by technical and methodological challenges such as background noise, signal variability, inter-speaker differences, and device heterogeneity.
Despite their significant promise, vocal biomarkers are not yet widely adopted in clinical research or routine care. Key barriers include limited transdisciplinary collaboration, insufficient large-scale clinical evidence, regulatory uncertainty, lack of standardization in data collection and implementation, and ethical and privacy concerns related to voice data sharing.
To overcome these challenges, coordinated collaboration among researchers, healthcare professionals, industry, patients, and regulatory bodies is essential. Establishing standardized, evidence-based, and ethically sound frameworks will be critical to enabling the integration of vocal biomarkers into healthcare. With such frameworks in place, vocal biomarkers have the potential to significantly improve disease diagnosis, monitoring, and treatment across a wide range of health conditions.
• Objectives
Strategic objectives for Vocal Biomarkers
Research Coordination
- To foster collaborative and interdisciplinary efforts in vocal biomarkers research, in order to bring together experts and expertise across different disciplines involved in vocal biomarker research
- To enhance standardisation and validation of vocal biomarkers by comparing and harmonising methods and approaches to design and validate vocal biomarkers, to go beyond the existing proofs of concept and increase the impact on the healthcare sector
- To ensure ethical, legal, and secure handling of personal data by identifying, addressing and mitigating ethical, data protection, and privacy issues that may arise in the collection, storage, use, and sharing of personal or sensitive data
- To disseminate vocal biomarkers research to stakeholders, to increase vocal biomarkers research impact
- To integrate vocal biomarkers with other digital therapeutics and explore the multimodal fusion with other technologies (e.g. video, wearable sensors, IoT, etc.) and relevant clinical parameters
Capacity Building
- To create and maintain a comprehensive repository of open source data and software resources for vocal biomarkers, accessible to all 3
- To foster establishment of large-scale, multilingual, longitudinal and privacy- preserving audio databases by defining the foundations and guidelines for building and secure sharing of audio databases accompanied with clinical data, to support vocal biomarkers research
- To establish an European framework to develop safe and trustworthy vocal biomarkers by creating guidelines for regulators, including those involved in AI, medical devices and the European Medicines Agency (EMA)
- To support knowledge transfer, inclusiveness, leadership development and interdisciplinary research