“The app analyses the different types of frequencies that are in the cry and the different patterns of sound and silence,” said Ariana Anderson, PhD, assistant professor in residence at the UCLA Semel Institute who spearheaded the development of the app.
“For example, if a cry has a long period of silence, it’s more likely that the baby is fussy. If there are constant, high-volume frequencies, it’s more likely the baby is in pain.”
Anderson is a mother of four who got the idea for the app after noticing the variations of cries in her own children. She then collected a database of more than 2,000 infant cries and used machine learning to build the algorithms used in the app.
“Current technology will alert parents when there is a sound coming from a child, but it doesn’t distinguish what type of sound it is,” said Anderson. “This categorises the cries to tell parents whether the baby is hungry or fussy or, with more than 90% accuracy, can determine if a baby is crying because it’s in pain”.
Discover more here.
Image credit: UCLA.