Zana Presents VOICE-BIOME Platform at 3rd Graz–Vienna Speech Workshop: Connecting Speech and Health Sciences

Vienna, April 25, 2025

Zana was selected to present at the 3rd Graz–Vienna Speech Workshop: Connecting with Health Sciences, hosted by the Medical University of Vienna from April 23–25, 2025. The event brought together top minds at the intersection of speech science, clinical research, and machine learning.

At the workshop, Zana presented VOICE-BIOME, its scalable and deployment-ready voice biomarker platform for cardiorespiratory health monitoring. In a compelling session led by Zana’s machine learning engineer Prashanth Pombala—a multidisciplinary expert in mathematics, physics, and electronics engineering—the team highlighted how voice data can be leveraged for passive, remote health screening at scale.

“Our goal with VOICE-BIOME is not just to build another algorithm,” said Prashanth, “but to enable reproducibility, cross-study comparability, and seamless integration with clinical workflows.”

VOICE-BIOME is designed as a modular, end-to-end system supporting data collection, acoustic feature processing, model retraining, and result visualization. It emphasizes adaptability to diverse research contexts and clinical applications, enabling researchers and health professionals to harness voice as a meaningful, non-invasive health signal.

VOICE-BIOME end-to-end Speech-based Health Monitoring Framework: A scalable platform for collecting, processing, and analysing patient speech data to detect health conditions. The system integrates real-world data collection, feature extraction, biomarker analysis, and secure data management to provide actionable health insight.

Throughout the conference, Zana engaged in lively cross-disciplinary discussions on the future of speech-based AI in healthcare, including the real-world deployment of voice biomarkers, machine learning scalability, and ethical considerations in speech technology.

The workshop themes that resonated most with us highlighted the growing intersection between speech science and healthcare. Sessions explored the use of voice as both a diagnostic and assistive tool across various health contexts, emphasizing its potential in clinical settings. From phonetics research to the development of intelligent speech interfaces, the event showcased how foundational linguistics is evolving alongside cutting-edge AI.

Hands-on innovation was a core part of the experience, with practical workshops on speech clustering, ultrasound imaging, and other tools that bridge academic research and real-world application.

Several breakthrough moments stood out:

• The launch of new AI-driven voice conversion initiatives aimed at supporting individuals with speech pathologies

• Notable progress in automatic speech recognition (ASR) for handling disfluent, spontaneous, and low-resource speech

• Research unveiling how noisy speech environments affect cognitive attention, opening new directions in clinical and user experience studies

• A strong focus on building clinical-grade tools for conditions such as Alzheimer’s, dysarthria, and vocal prosthetic use

Altogether, the event underscored the growing momentum behind speech-based technologies in health, aligning deeply with our mission at Zana.

Zana’s participation reaffirms its commitment to bridging cutting-edge AI with meaningful healthcare impact. Events like the Graz–Vienna Speech Workshop are vital in catalyzing new collaborations and moving the field closer to equitable, accessible, and personalized voice-enabled health technologies.