A team of researchers from the Indian Institute of Technology (IIT Guwahati), in collaboration with The Ohio State University, USA, has created a groundbreaking underwater vibration sensor that allows for contactless and automated voice recognition. This innovative technology is aimed at assisting individuals with voice disabilities who cannot rely on traditional voice-based systems.
Published in Advanced Functional Materials, the study was led by Prof. Uttam Manna from the Department of Chemistry at IIT Guwahati, alongside research scholars Debasmita Sarkar, Rajan Singh, Anirban Phukan, Priyam Mondal and Prof. Roy P. Paily from the Department of Electronics and Electrical Engineering. They were joined by Prof. Xiaoguang Wang and Ufuoma I. Kara from The Ohio State University.
The team noted that although voice recognition features are increasingly common in modern devices, they remain out of reach for many people with vocal impairments. Research indicates a significant number of children and young adults aged 3 to 21 suffer from voice-related challenges, underlining the importance of accessible alternatives.
To bridge this gap, the researchers explored how the airflow generated during speech without actual sound can be used for recognition. When someone attempts to speak, even in silence, the breath they release creates gentle ripples on a water surface. The team’s sensor, placed just beneath the air-water boundary, captures these minute vibrations and converts them into electrical signals.
Constructed using a conductive, chemically sensitive porous sponge, the sensor works with AI-powered Convolutional Neural Networks (CNNs) to decode these signals and identify attempted speech. This allows users to interact with devices in a completely hands-free and soundless manner.
The current lab prototype was developed at a cost of approximately ₹3,000, and further efforts are being made to lower this cost, potentially with help from industry partners. The researchers highlight that the sensor has demonstrated long-term durability in underwater conditions and may also be useful in other areas like fitness tracking, motion sensing and underwater monitoring.
Looking ahead, the team plans to conduct clinical trials and gather more data from individuals with speech impairments to fine-tune the system’s ability to detect specific words and commands used in smart technologies.
Commenting on the development, Prof. Uttam Manna said,
“This is one of the rare material designs that enables voice recognition by observing the water waves created by exhaled air at the air-water interface. It holds great promise as a communication tool for people with damaged or non-functioning vocal cords.“
