Apr 16, 2026 | Read time 2 min

AI can now understand health signals from 15 seconds of your voice, including fatigue, stress and type 2 diabetes

The joint platform returns transcription and health signals in real time, with no additional hardware required.
Speechmatics x Thymia - Header (1200 × 900)
Speechmatics
SpeechmaticsEditorial Team

Speechmatics and thymia are combining medical-grade speech-to-text with clinical-grade voice biomarker intelligence to identify health signals carried in every voice conversation. 

Their joint platform identifies 30-plus health signals from 15 seconds of natural speech, including stress, fatigue, depression and anxiety symptoms, type 2 diabetes and driver impairment, then acts on them in real time.

Delivered through a single integration, the offering returns health signals alongside standard transcription at over 85% accuracy. No additional hardware is needed. 

15 seconds of audio, 30 health signals 

thymia’s medical technology processes audio at 15-second intervals through a single integration, reading neurological signatures across both acoustic and linguistic patterns to extract health signals from natural speech.  

Built on the largest dataset of its kind, including 75,000+ unique voices measured over time, thymia's science is validated in 20-plus peer-reviewed publications including Nature Scientific Reports. Its contextual interpreter then translates those signals into actionable recommendations. 

Speechmatics' speech engine, trained on millions of hours of audio, handles accuracy across accented speech and non-native speakers. That breadth matters: a technology like this, deployed at population scale, has to work equally well regardless of how someone speaks. 

thymia has built something that turns an everyday voice interaction into actionable health insights. Our job is to make sure that works for everyone who speaks, not just the people whose voices AI has historically been built around. That’s the version of this technology worth building.” Katy Wigdahl CEO, Speechmatics 

 

A $5 billion market and a July 2026 EU mandate 

Six years of research and development underpin thymia’s technology, with research published in Nature Scientific Reports. Built on rigorous science, one of the largest and most diverse datasets of its kind, and safety-first design, the company is now pursuing medical device regulatory approvals — the final validation of a technology already trusted in safety-critical industries. 

Beyond healthcare, thymia's technology is already deployed in safety-critical industries including automotive, with applications spanning voice AI agents, contact centers and telecoms. Technology like thymia’s is fast becoming essential: the EU mandates driver fatigue monitoring in all new vehicles by July 2026, and the voice biomarker market is projected to reach $5 billion by 2028. 

Until now, no technology could read the complete signal carried in a voice conversation, at scale. Together with Speechmatics, we combine medical-grade speech-to-text with clinical-grade voice biomarker intelligence, understanding not just what is said but how the human saying it actually is. Emilia Molimpakis CEO & Co-founder, thymia 

 

Launching at Voice AI Space 

The partnership will be launched this weekend at THE VOICE AI HACK, a hackathon centered on medical applications of voice AI. Developers and healthcare innovators are invited to build on the combined Speechmatics-thymia platform. Register at: luma.com/voiceaihack

About Speechmatics 

Speechmatics is the Voice AI company on a mission to understand every voice. Its speech-to-text technology delivers industry-leading accuracy across 55+ languages, with specialized models for healthcare, media, contact centers, and enterprise organizations worldwide. Speechmatics powers leading technology providers including Adobe, AI Media, Content Guru, and Nordhealth, and offers deployment across cloud, on-prem, and on-device. www.speechmatics.com

About thymia 

thymia is the human state voice intelligence company. Its technology reads the complete signal carried in voice, detecting 30+ clinical-grade biomarkers from just 15 seconds of speech, and acts on it in real time, delivering safety and experience value to partners. Built on one of the largest datasets of its kind, including 75,000+ unique voices, thymia is trusted across healthcare, safety-critical industries and beyond. Learn more at thymia.ai  

Media contacts

Mieke Smith, Speechmatics: mieke.smith@speechmatics.com  Stella Lempidaki, thymia: info@thymia.ai