MIT Speech recognition software flushes out depression. “It’s not what you say, it’s how you say it,” says . . .
“It’s not what you say, it’s how you say it,” says group at MIT’s Media Lab. They believe it’s the tone and pitch of a person’s voice as well as the length and frequency of the pauses combined with speed of speech speed, that reveals true mood. This isn’t the first time man has implemented speech recognition software for greater insights into our bodies and minds. A few years ago, the pharmaceutical giant Pfizer developed voice-analysis software that detected early signs of Parkinson’s disease, employing highly sophisticated computer algorithms they were/are able to successfully recognize tiny tremors in speech which flushed out clues which helped them gauge patients’ responses to various medications. Now, Professor Sandy Pentland at MIT’s Media Lab, is developing algorithms with the intention of analyzing subtle cues in speech that indicate awkward, anxious, disconnected or depressed feelings. For years, psychiatrists have recognized a characteristic pattern in the way that many people with clinical depression speak: slowly, quietly and often in a halting monotone. One company, Cogito Health, spun out of MIT, is building on Pentland’s research by developing voice-analysis software to screen for depression over the phone. The way the system is currently managed, nurses make routine calls to patients between visits to ask if they are taking their medication. This is fine, but symptoms of depression are often too subtle for nurses to readily identify. Voice analysis software could provide a natural and noninvasive way for nurses to screen for depression during routine phone calls.
“If you’re a nurse and you’re trying to deal with a patient with long-term diabetes, it’s very hard to tell if a person is depressed,” says Joshua Feast of Cogito Health. “We try to help nurses detect possible mood disorders in patients that have chronic disease.”