A team of researchers at the Indian Institute of Technology (IIT) Guwahati has developed an underwater vibration sensor that enables automated and contactless voice recognition. The novel sensor developed by IIT Guwahati in collaboration with researchers from Ohio State University offers a promising alternative communication method for individuals with voice-based disabilities.
How does the sensor work?
The Rs 3,000 sensor works on a very simple technique. It focuses on the air exhaled by a person through the mouth while speaking – a basic physiological function that is exercised by almost all humans.
In cases where individuals cannot produce sound, attempting to speak still generates airflow from their lungs that then exits from their mouth. When directed over a water surface, this air flow produces subtle waves and patterns on the water.
“This novel underwater vibration sensor could detect these impressions made on the water by the airflow to interpret and produce speech signals without depending on audible voice, thus creating a new pathway for voice recognition,” the researchers behind the study said.
The research was published in the Journal of Advanced Functional Materials in November last year. Titled ‘Underwater Vibration Sensor to Enable Automated and Contactless Voice Recognition’, was co-authored by research scholars Debasmita Sarkar, Rajan Singh, Anirban Phukan, and Priyam Mondal, along with Prof Roy P Paily from the Department of Electronics and Electrical Engineering, IIT Guwahati.
“It is one of the rare designs of material allowing the recognition of voice based on monitoring the water wave formed at the air/water interface because of exhaling air from the mouth. This approach is likely to provide a viable solution for communication with those individuals with partially or entirely damaged vocal cords,” said Prof Uttam Manna, Department of Chemistry, IIT Guwahati.
When placed just below the air-water interface, the new sensor captures the tiny disturbances created by exhaled air and converts them into measurable electrical signals.
The research team utilised Convolutional Neural Networks (CNNs), a deep learning model, for accurate recognition of subtle signal patterns. This configuration enables users to communicate with devices remotely, eliminating the need for sound generation.
“On a lab scale, the working prototype costs Rs 3,000,” the research paper mentions, noting that they are exploring potential industry collaboration for bringing the technology from the lab to real-world use, which can reduce the cost.
What are some of the other features of the sensor?
Besides aiding individuals with voice disabilities, this sensor also demonstrates potential in areas like hands-free control of smart devices, movement detection, exercise tracking, and even underwater communication.
Beyond voice recognition, the sensor can also be utilised for exercise tracking and movement detection. In addition, its proven durability, remaining stable after extended underwater use, suggests potential applications in underwater sensing and communication.