New App Senses Emotions From Voice Tone
Remember mood rings? With their thermochromic elements, such as liquid crystal, they changed color in response to the wearer’s body temperature, supposedly to indicate his or her emotional state. Now, the developers of a new smartphone app claim it can also predict a user’s mood by monitoring changes in voice tone.
The Xpression app—developed by El Technologies’ Matt Dobson and Duncan Barclay—specializes in speech recognition software. Once installed, it remains active in the background and “listens” to the user’s voice on all calls made on the phone, tracking voice loudness, intensity, pitch and speed—factors that can point to five emotional states: Calm, happy, sad, angry and anxious. The emotions are then noted by the app and added to a list tracking the user’s moods and when they change. The list can then be emailed to third party—such as a psychologist—at the end of each day.
Adrian Skinner, a clinical psychologist at the UK’s National Health Service, told New Scientist there is a strong need for such technology. Patients with depression and anxiety are often asked to record their mood changes throughout the day in order to help psychologists fine-tune their treatments. “With conditions like depression, people tend to stop doing things like filling in mood diaries,” Skinner said. “If this app gives us more complete diaries it could help us better find the day-to-day triggers that raise or lower a patient’s mood.”
How does Xpression work? It sends 200-millisecond-long acoustic snapshot to a remote server where the results are matched against a database to estimate the user’s likely emotional before sending it back to the app for storage. For privacy sake, actual calls are not recorded and no central data about the user is collected.
The Xpression app is not available in the App Store just yet, however. El Technologies is currently working to build interest among interest among insurance companies and governments. One British company, for example, is reportedly looking into using the app to track workplace stress levels to measure the effectiveness of stress therapy offered to employees. Clinical trials are scheduled for later this year.
[Image via Shutterstock]
I want more stuff like this!