An Hour of Innovation with Vit Lyoshin The Future of Medicine! AI That Listens to Your Voice for Early Disease Detection | Henry O’Connell
In this episode of An Hour of Innovation podcast, host Vit Lyoshin sits down with Henry O’Connell, CEO and Co-Founder of Canary Speech, to explore how artificial intelligence is transforming healthcare through the power of the human voice.
Henry shares the fascinating story of how a simple conversation in a bagel shop led to the creation of a breakthrough technology that analyzes how we speak, not what we say, to detect early signs of neurological and mental health conditions such as Parkinson’s, Alzheimer’s, anxiety, and depression. We discuss how AI voice biomarkers work, how Canary Speech can process 15 million data points per minute, and how this innovation enables early disease detection through non-invasive, real-time speech analysis.
Henry also explains how this technology integrates with wearables and clinical tools, allowing doctors to monitor patients through their voice using smartphones, smartwatches, and telemedicine platforms. He reveals how machine learning models can recognize patterns invisible to the human ear, helping medical professionals diagnose faster, more accurately, and with greater empathy.
Beyond the technology, Henry shares his vision for the future of medicine, his passion for accessibility in global healthcare, and his lessons from building a company at the intersection of AI, neuroscience, and digital health innovation.
Henry O’Connell is a healthcare innovator and entrepreneur. As the CEO of Canary Speech, he leads a team pioneering the use of voice biomarkers for early detection of diseases that affect millions worldwide. His work has been published in The Lancet and recognized by global partners including Microsoft, Samsung, and Harvard’s Beth Israel Hospital. Under his leadership, Canary Speech is redefining how AI can listen, understand, and improve human health.
Support This Podcast
* To support our work, please check out our sponsors and get discounts: https://www.anhourofinnovation.com/sponsors/
Takeaways
* AI can now analyze how we speak, not what we say, to detect health conditions.
* The human voice contains over 2,500 measurable biomarkers tied to brain and body function.
* Canary Speech analyzes 15 million data points per minute from spoken language.
* Diseases like Parkinson’s, Alzheimer’s, anxiety, and depression can be detected early through voice.
* The central nervous system controls speech, making it a powerful source of diagnostic data.
* Voice biomarkers are universal; the same disease patterns appear across languages and cultures.
* The technology works through smartphones, telehealth systems, and even Samsung watches.
* Canary Speech partners with leading institutions like Harvard, Hackensack Meridian, and The Lancet.
* AI enables non-invasive, real-time health monitoring during normal conversations.
* The system can even detect stress, fatigue, and aggression, helping prevent hospital incidents.
* Voice-based diagnostics may one day redefine patient-doctor interactions.
Timestamps
00:00 Introduction
01:56 The Birth of Canary Speech
07:00 Can AI Hear Illness?
14:25 Inside Voice Biomarkers
22:21 Detecting Anxiety, Depression, and Cognitive Decline Through Speech
25:07 How AI Voice Technology Works in Real Healthcare Settings
26:29 Preventing Aggression and Protecting Care Teams
29:56 Giving Doctors a New Diagnostic Superpower
31:10 What Doctors Say?
32:17 Everyday Applications of Voice AI
34:57 Making AI-Driven Healthcare Accessible Worldwide
36:11 Data Privacy, Ethics, and Trust in AI Healthcare
38:49 Could We Talk to Animals?
42:00 Leadership, Failure, and Building with Purpose
Connect with Henry
* Website: https://canaryspeech.com/
* LinkedIn: https://www.linkedin.com/in/henryjoconnell/
Connect with Vit
* Website: https://vitlyoshin.com/contact/
* LinkedIn: https://www.linkedin.com/in/vit-lyoshin/
* X: https://x.com/vitlyoshin
Vit’s Projects
* Podcast: https://www.anhourofinnovation.com/
* AI Assistant to build apps: https://appforgelab.com/