My research interests focus on Computational Neuroscience, Speech Processing, and Multilingual NLP, with a core goal of understanding how AI models align with the human brain.
I have worked on: [1] identifying speech markers for depression, to build tools for mental health; [2] developing prosody based models that use pitch and tone to better understand the hidden meaning in spoken language; [3] handling multilingual and codemixed speech (mixing languages like Hindi and English) for more accurate sentiment analysis; and [4] studying the architectural biases of different models when processing encrypted text across various languages. Through this work, I hope to create speech and language technology that is more human centric and socially impactful.