Strategy& Insider Podcast - Episode 46 with Lara Gervaise and Edoardo Guidice

Why It Matters

By turning everyday speech into a medical biomarker, Vuosis AI could enable mass‑scale, low‑cost early detection, reshaping preventive healthcare and reducing the burden of chronic disease.

Key Takeaways

• Voice AI can detect up to 25 health conditions from 30 seconds
• Early detection includes Parkinson years before symptoms appear
• Accuracy varies: 96% for Parkinson, ~80% for mental health
• Technology is language‑agnostic, focusing on paraverbal acoustic features
• Backed by Microsoft and EPFL spin‑off, targeting global health market

Summary

The Strategy& Insider podcast featured Lara Gervaise and Edoardo Guidice, co‑founders of Vuosis AI, a Swiss EPFL spin‑off that uses voice analysis to flag early signs of fatal diseases, burnout and cognitive decline.

Vuosis AI’s platform extracts hundreds of acoustic features—tone, intonation, rhythm—from a 30‑second natural speech sample and runs them through AI models trained on healthy and diseased cohorts. The system can assess up to 25 conditions, from neuro‑degenerative disorders such as Parkinson’s and Alzheimer’s to mental‑health issues and even cardiac or metabolic diseases.

Lara highlighted that the technology can detect Parkinson’s up to seven years before tremors, with 96% sensitivity and specificity, while depression detection sits around 77‑83%. Edoardo emphasized the rapid AI development cycle—three months comparable to three years in other tech—and the model’s language‑agnostic, paraverbal design that preserves privacy.

If scaled, the tool could shift diagnostics from invasive tests to passive, ubiquitous monitoring, lowering costs and expanding early‑intervention opportunities worldwide. Partnerships with Microsoft and interest from health insurers suggest a fast‑moving market, though challenges remain in multilingual validation and regulatory approval.

Original Description

Lara Gervaise and Edoardo Giudice are the co-founders of Virtuosis AI, a start-up revolutionizing early disease detection through voice biomarker technology. In this episode, they explain how just 30 seconds of natural speech can be analyzed to screen for up to 25 health conditions – from Parkinson’s and Alzheimer’s to depression, anxiety, and cardiometabolic disorders. They discuss their journey from a chance meeting before generative AI existed to developing a CE-certified medical device designed to transform preventive healthcare at scale. The conversation also covers the ethical safeguards involved in workplace integration, the challenges of building health AI within Europe’s regulatory framework, and their vision for a future where voice monitoring is as routine as tracking your steps.

Tune in for an inspiring discussion on the intersection of AI, voice technology, and the future of preventive care.

To learn more about the Strategy& Insider Podcast, visit

https://www.strategyand.pwc.com/de/en/insider-podcast.html

Explore our latest insights on healthcare innovation and digital transformation here:

https://www.strategyand.pwc.com/de/en/industries/pharma-life-sciences.html