IISC Lab enhances hearing technology with the help of AI
In one test, people had to press a button when they noticed a change in the speaker. Interestingly, those who didn't understand the language were quicker to identify changes. “When language is unknown, the brain pays more attention to the voice and tone, not the meaning,” explained Ganapathy, saying this helps design smarter hearing aids.
Most of Ganapathy's work is “maintainer learning,” with machines being trained to pick patterns from the audio. These patterns help the machine identify different voices, accents and emotions – even when the speech is less clear. Aside from better hearing aids, this study will help build virtual assistants that sound more human.
The lab is also working on explainable AI (XAI). This analyses how reliable AI decisions are. In one project, the lab trained an image recognition model to identify important visual areas mentioned in the caption. It performed better than the older models. This can also be used in voice models, and AI can help you understand and explain the answer more clearly.
During Covid-19, Ganapathy's team created Coswara. Coswara is an app that detects possible COVID infections using cough and voice sounds. They trained the models using a sample of people across India. The Indian Medical Council (ICMR) discussed expanding it, but it did not move forward. The team later released the survey results in 2023. His journey began in Kerala, where he studied electronics and communications at the Faculty of Engineering at Trivandrum (CET). He later joined IISC and participated in the MTECH of signal processing. .
Everything changed when I attended a lecture by Hynek Hermansky at the Swiss EPFL (Swiss Federal Institute of Technology). Ganapathy contacted him, where he joined his lab. After his PhD he worked at the IBM TJ Watson Research Center.
