Saed News: An artificial intelligence development startup has used machine learning technology to translate babies’ cries, aiming to help AI bring parents and their infants closer together.
According to SAEDNEWS, citing ISNA, newborns express their needs through different types of crying, but sometimes parents are unable to accurately understand those needs and require assistance in communicating with their child.
Now, according to a Switzerland- and Spain-based startup called Zoundream, this help may come in the form of artificial intelligence and a dedicated device that translates what they call the “universal language of babies.”
Roberto Iannone, CEO of the startup, said in an interview at the Mobile World Congress in Barcelona: “Every time a baby cries, it wants something. Our company’s idea is based on the principle that children cry in different ways for different needs.”
Although placing the words “baby” and “AI” together may sound like science fiction, the device appears to work well. However, Iannone emphasizes that it is by no means a replacement for parents’ intuition and only acts as a supportive tool to improve their ability to understand the baby’s needs.
He said: “Our device is absolutely not designed to replace parents in understanding their babies’ needs, but rather to increase their confidence and awareness of how the baby is crying and what it needs.”
Zoundream used machine learning software to analyze thousands of hours of international baby cries and classified them into four categories: hunger, pain, gas, and the need for comfort (being held).
The company was founded by Iannone together with a data scientist named Anna Laguna.
Iannone explained: “When we first started, we bought cheap audio recorders and asked some parents—and even paid some of them—to record their babies’ sounds closely.”
He said: “It was a very time-consuming process, but at least it was a start. Then, as soon as it became possible, we developed our own device, basically a recorder that filtered out all sounds except baby cries.”
He added that this was good for privacy and also meant they no longer had to manually filter long hours of irrelevant sounds. This method provided important data.
He also emphasized the role of “prosody” in early infant development, which is the ability to recognize and communicate through melodic tones rather than speech. This ability begins developing even in the womb. Iannone said this “language of the baby itself” plays a key role in the company’s research.
The next step is bringing the device to market, which Iannone says will happen by this summer.
Regarding privacy concerns and whether parents would accept the technology, he said: “Parents love technology, and usage depends on their needs. Everyone likes it for different reasons.”
He added: “Generally, it’s not that parents have no idea why their baby is crying. But sometimes they are unsure, especially first-time parents. In other cases, parents think they know what the baby wants, but still want confirmation.”
