AI can help you figure if your baby is fussy, hungry or in pain with machine-learning technology

Artificial intelligence (AI) has made significant strides in everyday applications such as speech recognition and machine translation. These technologies allow machines to comprehend our words and convert them into actionable commands.

Now, a new application of AI aims to interpret a baby’s cries, providing insights into what they are trying to communicate. Researchers at the University of California, Los Angeles have developed a free app called Chatterbaby for Android and iOS devices.

The app utilises an algorithm that can interpret different types of baby cries and relay this information to parents. The creators claim that the app can do this with remarkable accuracy.

https://t.co/32wsdRSdt5 How to Understand Baby Sounds
The ChatterBaby app developed by @UCLAHealth Dr. Ariana Anderson helps worried parents interpret cries.

— UCLASemelFriend (@UCLASemelFriend) March 3, 2022

The Inspiration Behind Chatterbaby

Ariana Anderson, assistant professor and lead researcher at UCLA, was inspired to create the app after noticing similarities in the cries of her three children. As a statistician, she saw patterns in these cries and wanted to test if these patterns were present in other children as well.

To develop Chatterbaby, Anderson and her team uploaded 2,000 audio samples of infant cries. They then used AI algorithms to distinguish between cries induced by pain, hunger, and fussiness. The training involved extracting acoustic features from their database of pre-labeled cries.

Pain cries were taken during vaccinations and ear piercings, while other cries were labelled using parent-nomination and a ‘mom-panel’ consisting of veteran mothers who had at least two children. The algorithm was trained using these acoustic features to predict the most likely reason for the cry.

Within their sample, the algorithm was about 90 percent accurate in identifying pain cries and over 70 percent accurate overall.

A new app could help parents interpret why their baby is upset-- based on how they cry. The free app "Chatterbaby" analyzes the acoustic features of a baby's cry, to help parents understand whether their child might be hungry, fussy or in pain: https://t.co/jwCsmQsh4N pic.twitter.com/BstJKW97zQ

— Local 12/WKRC-TV (@Local12) July 2, 2018

However, Anderson emphasises that parents should still rely on their instincts as they are more powerful than any AI algorithm.

Potential uses & benefits of Chatterbaby

Chatterbaby could be particularly useful for first-time parents or those who are deaf or hard of hearing. It could also be a valuable tool in diagnosing autism at an earlier age. Currently, autism is typically diagnosed around the age of three.

Early detection could allow for earlier intervention. Anderson suggests that unusual vocal patterns in infants could be an early indicator of autism.

By inviting people into our research study with the Chatterbaby cry translator, we can them follow them for six years and provide free screenings for autism which they can do from their own home

Ariana Anderson, assistant professor and lead researcher at UCLA

While previous studies have shown promising results in detecting abnormal vocal patterns in at-risk children, these studies have small sample sizes. To gather more data, Chatterbaby offers a voluntary study for parents to participate in. Although it’s still in its early stages, it could potentially provide valuable insights for earlier diagnosis in the future.