A Singapore-based provider of a digital healthcare platform is looking into the potential of speech recognition and voice biometrics to improve healthcare services.
Called Doctor Smart, the company has teamed up with a language technology specialist called LangNet in what it calls a “first step” toward its voice-based research. Part of the aim is to alleviate the administrative burden on doctors through the use of speech recognition, which can be used to transcribe their notes and record patient stories, and even to streamline diagnosis through AI analysis of those texts.
This kind of use of speech recognition technology is increasingly common in healthcare, with major companies like Nuance Communications and Philips offering their own solutions. But Doctor Smart and LangNet have even greater ambitions concerning voice biometrics, asserting in a statement that such technology “has shown promising results for disease diagnosis,” and noting that a recent Mayo Clinic study found a link between voice characteristics and coronary artery disease. Elaborating on the possibilities, Doctor Smart co-founder Pavel Roytberg explains that “AI and machine learning could find vocal patterns that might signal stress disorder, postpartum depression, which is widely believed to be underdiagnosed, older people with dementia, Parkinson’s, coronary artery disease and others,” adding that monitoring changes in patients’ voices could alert doctors to times when they have stopped taking certain medications.
It’s an ambitious idea, but one that could pay off in a big way as AI and biometric technologies increasingly find their way into clinical settings. For now, though, Doctor Smart and LangNet’s collaboration is just getting started.
July 30, 2018 – by Alex Perala