Worked on applying Machine learning and signal processing techniques on real-world voice data in order to improve existing product offerings.

Areas of work: Speech processing and Machine Learning

Built a model which aims to determine the language in a given speech signal from dataset consisting of seven closely related Indian languages.

Developed a model to detect Indian English accent from a mix of all English speech.

Worked with voice activity detection, Gaussian Mixture Model (GMM), I-vector techniques, Artificial neural networks and Support Vector Machines

Areas of work: Machine Learning and Image Processing

Built a model for segmenting the smart-phone captured image of the wound into its correspond tissue types- necrosis, sloughy and granular tissue. The model involved extraction of color and texture features and training a classifier. The random forest classifier and Support Vector Machine Classifier were used.