A mandatory science fair assignment at Challenger School’s Strawberry Park campus in West San Jose has evolved into a significant invention for 14-year-old Arjun Goli of Saratoga. Goli has spent approximately two years creating an artificial intelligence-powered hearing aid designed to assist older adults with hearing difficulties.
The project began when Goli, a middle school student, sought to address a challenge within his own family. His grandparents, both in their 70s, experienced trouble hearing, prompting him to build a device that could overcome that barrier. His initial prototype functioned as a speech-to-text machine, utilizing a microphone connected to hardware that processed audio and displayed transcribed text on a screen or phone.
The school community responded positively to the early version, encouraging Goli to submit it to the 2025 Synopsys Science & Technology Championship. He secured second place in the physical science and engineering category. This recognition shifted his perspective on the project’s value, leading him to pursue it with greater seriousness rather than viewing it as a simple school requirement.
To advance the technology, Goli applied for a grant through Lead4Change.org, a leadership program supported by businessman David Novak. The program facilitated feedback sessions with residents at Morning Star Senior Living in West San Jose. His first presentation to five to eight seniors received lukewarm reactions, with one resident stating the design would not be helpful. Goli described feeling disappointed by the response but recognized the value of the critique in identifying necessary improvements.
Following the feedback, Goli refined the device to create a second version. This iteration aims to amplify a specific speaker’s voice in noisy environments by combining neuroscience with computer science. The system uses machine learning models and artificial intelligence for auditory attention decoding. It detects tiny electrical signals emitted by the brain when a person focuses on a particular sound, using sensors on an electroencephalogram to identify which speaker the user intends to hear.
Goli successfully secured $10,000 in seed funding for the project after implementing these changes. He returned to Morning Star Senior Living to present the updated device to a group of 20 to 30 older adults, who responded much more favorably than during the initial pitch. The invention melds wearable hardware with algorithms that Goli coded himself, utilizing machine learning capabilities to enhance auditory focus rather than simply increasing volume.