Speech scientists are developing a system that can monitor fatigue levels by looking for signs of tiredness in the voice, with plans for its initial use by heavy construction vehicle drivers. Matthew Stock reports.
Subtle changes to your voice can reveal your fatigue levels, say speech scientists in London. They've developed an algorithm to extract information from a voice recording to tell when you're tired. Spotting fatigue like this could one day be a life-saver on the roads. In Britain almost 20 percent of traffic accidents are sleep-related. SOUNDBITE (English) MARK HUCKVALE, PROFESSOR OF SPEECH SCIENCE AT UNIVERSITY COLLEGE LONDON, SAYING: "What we're trying to do is build a model of a person's voice by making multiple recordings over a period of time... Previous work has tended to focus on single characteristics like pitch height and pitch range. We're trying to use a larger constellation of features which involve many different aspects of both pitch and voice quality, and prosody, and speaking rate, and pausing; and many more things. And this gives us increased sensitivity in looking at the way in which the voice changes through the day and with time." Working with fatigue management firm WOMBATT, they've conducted a pilot study on a mine in Peru. Here, fatigue can be particularly dangerous for drivers operating these huge mining trucks. Despite moving from a controlled lab environment to the noise of a vehicle cabin, the effect of fatigue on the driver's voice could still be detected. Eventually, researchers foresee an unobtrusive in-cabin system. SOUNDBITE (English) MARK HUCKVALE, PROFESSOR OF SPEECH SCIENCE AT UNIVERSITY COLLEGE LONDON, SAYING: "The goal is not to stop the driver doing their work or to prosecute them or to stop the truck driving; it's simply to protect both the health of the driver and of course the safety of the vehicle." A new study is underway to explore how fatigue level in the voice is affected by changing shift patterns. The team hopes it will lead to a commercial system based on voice analysis in the next few years.