Researchers at IIT-Guwahati, in collaboration with Ohio State University (USA), have developed a groundbreaking underwater vibration sensor that enables automated, contactless voice recognition. This innovation opens up new communication possibilities for individuals with voice disabilities who cannot use conventional voice-based systems.
Overcoming the Barrier of Voice Disorders
While voice recognition technology is now central to interacting with smartphones and smart appliances, it remains inaccessible to those suffering from voice disorders. To address this gap, the research team focused on a fundamental physiological function — the flow of exhaled air during attempted speech — rather than relying on audible sound.
How the Sensor Works
When individuals unable to speak exhale while trying to form words, the air creates subtle disturbances on a water surface. The newly developed sensor, placed just below the air-water interface, detects these tiny waves. Made from a conductive and chemically reactive porous sponge, the sensor converts the disturbances into measurable electrical signals.
AI-Powered Interpretation of Air-Wave Patterns
As reported by The Mint, the team trained Convolutional Neural Networks (CNNs), a type of deep learning model, to interpret these delicate signal patterns. This allows the system to recognize speech-like commands without needing the user to produce sound, enabling hands-free control of devices from a distance.
A Step Toward Inclusive Communication
Professor Uttam Manna, a key member of the research team, explained, “This is one of the rare material designs capable of recognizing voice through monitoring the water waves generated by exhaled air. It presents a promising communication method for individuals with partially or fully damaged vocal cords.”
Cost-Effective Prototype with Future Potential
Currently, the working prototype costs around ₹3,000 at the lab scale. However, researchers are actively exploring industry collaborations to bring the technology to market. Larger-scale production is expected to significantly reduce the cost.
Next Steps: Clinical Validation
The research team now plans to pursue clinical validation for the device to assess its real-world effectiveness and usability. The findings have already been published in the journal Advanced Functional Materials, signaling a promising future for assistive voice recognition technologies.
Key Features of the Sensor
- Contactless, soundless communication
- AI-enabled signal interpretation using CNNs
- Low-cost prototype
- Hands-free control of smart devices
- Potential solution for people with speech impairments
With this innovation, IIT-Guwahati and its partners are paving the way for a more inclusive technological future. This advancement enables everyone — regardless of their ability to speak — to interact seamlessly with smart systems.