Author: Stefani Kim

Preventing AI Audio Models From Being ‘Fooled’

Researchers at the UPV/EHU-University of the Basque Country show that the distortion metrics used to detect intentional perturbations in audio signals are not a reliable measure of human perception, and have proposed a series of improvements. These perturbations in audio signals, designed to be imperceptible, can be used to cause erroneous predictions in artificial intelligence (AI).

Read More

Brain Injuries In Utero May Impact Language

A new UC San Francisco (UCSF) study that mapped the neural connections of newborns with two different kinds of brain injuries found the maps looked very different – and were linked to significantly different developmental outcomes years later.

Read More

xMEMS Launches Montara Pro MEMS Microspeaker

xMEMS Labs announced the launch of its Montara Pro, described as “the world’s first monolithic MEMS µspeaker with integrated DynamicVent enabling smart TWS earbuds and hearing aids that create best-of-both-worlds’ user experiences combining the benefits of closed-fit (occluded) and open-fit earbuds.”

Read More

Children with ASD May Respond Less To ‘Motherese’ Speech

Previous research has shown that infants prefer to listen to motherese, more formally known as infant-directed speech, over adult-like speech; that it more effectively holds their attention and is an important component of emotional bonding and fosters learning experiences between child and parents.

Read More

Mapping the Musical Mind

Researchers in Japan used magnetic resonance imaging to study the brains of secondary school students during a task focused on musical observation. They found that students trained to play music from a young age exhibited certain kinds of brain activity more strongly than other students. The researchers also observed a specific link between musical processing and areas of the brain associated with language processing for the first time.

Read More

Research Shows 1,500 Languages At Risk By Century’s End

A “world-first study” warns 1,500 endangered languages could no longer be spoken by the end of this century. The study, led by The Australian National University (ANU), identified predictors that put endangered languages at high risk. An article detailing the research appears on the University’s website.

Read More