Today we’re talking about something surprisingly revealing, your voice. Not what you say, but how you say it.

Because according to Canary Speech, a 20-second audio clip could tell more about your mental health than a half-hour questionnaire.

It’s like having a stethoscope for stress, anxiety and even cognitive decline, except instead of checking your heartbeat, it listens to your “How are you?” and goes, “Actually… you don’t sound fine.”

In this conversation, Henry O’Connell, Co-Founder and CEO of Canary Speech, helps us understand how machine learning is turning everyday speech into a new kind of health indicator, and whether that’s brilliant, slightly creepy or a bit of both.