Zoundream harnesses the power of artificial intelligence (AI) and advanced sound recognition technology to decode and interpret the cries of newborns. This innovative approach focuses on the specialized realm of baby cries, enabling real-time translation of various needs such as hunger or fatigue.

However, the primary objective goes beyond mere translation. Zoundream aims to facilitate the early detection and diagnosis of potential health issues and developmental disorders in newborns. By doing so, It endeavours to enhance the well-being of families and infants worldwide.

Zoundream: unveiling the meaning of baby cries

Globally, approximately 130 million babies are born annually, with 47 million of these births occurring in developed nations. In these regions, new parents typically allocate around $10,000 per child in the first year alone, a steadily rising figure. This trend underscores the burgeoning growth of the parenting industry, which boasts a Compound Annual Growth Rate (CAGR) of approximately 37% and is poised to soar to an estimated $81 billion by 2025.

Baby cries transcend mere sounds; they serve as a nuanced form of communication, conveying vital information about a newborn’s needs, emotions, health, and overall well-being. Particularly for infants, crying constitutes their primary mode of interaction with the world around them.

AMSI – the platform

AMSI is now a mature platform that Zoundream offers to partners. “Initially, we prototyped AMSI on a hardware device that was used for testing purposes only. This constraint required us to optimize the AMSI engine for various potential architectures, including embedded hardware platforms and mobile systems. However, we have now migrated AMSI to a remote service-based solution to take advantage of the increased computational power of cloud computing and provide even more accurate results,” the company states.

AMSI has been designed to run on various hardware and platforms, potentially any connected device with a microphone. AMSI technology is tested and used by actual parents with their newborns. As of the beginning of 2021, Zoundream had totalled over 100,000 hours of testing—and counting.

“In Zoundream, we’ve spent several months to determine the best solution to identify and translate a human-produced sound into its actual meaning, specifically for children pre-speech. There are several approaches in Machine Learning literature to address this challenge, but unfortunately, none of them is suitable to work on a continuous incoming stream of audio. So we created our custom deep-learning-based solution. AMSI, which is an acronym for Acoustic MultiStage Interpreter, processes the sound into a series of inner deep-learning models to extract the meaning of a baby cry. Think about understanding letters, words, then meaningful sentences. That’s exactly what AMSI does. AMSI can be adapted to many different problems and shortly we’ll be able to identify pathologies and uncommon behaviors of the baby,” Zoundream underlines.

For small babies, crying serves as their sole means of communication with the world, constituting a natural and instinctive language. Remarkably, it is the universal language infants share worldwide, transcending gender, culture, or nationality; all babies consistently express their needs through cries. Through sound analysis, it becomes feasible to discern the underlying reasons behind a baby’s cries. However, viewing baby cries through the lens of communication is merely the initial phase of an innovative journey into the realm of cry analysis.