Projects

1. Age, Gender and Emotion Recognition through Speech Spectrograms using Feature Learning

   • Designed a state-of-the-art technique for identifying various speaker characteristics, including age, gender, and emotion, by analyzing sampled audio.
   • Analyzed the characteristics of the speech signal, such as MFCC, ZCR and RMS energy.
   • Evaluated system's accuracy, resulting in 79.57% and 93.26% accuracy for age and gender classification respectively, and 98% accuracy for emotion recognition.
2. e-Yantra Learning Analytics Dashboard (e-YLAD) 2.0

   • Built upon an earlier dashboard (eYLAD 1.0) for theme developers (instructors) of the e-Yantra Robotics Competition to help them monitor the status and activities of participating teams.
   • Worked on revamping the existing dashboard to improve the user experience by identifying better use cases and getting more insights.
3. Holonomic Art Robot

   • Assembled a Holonomic drive-based bot for deployment in an arena that is an abstraction of different settings in a smart city.
   • Developed path tracking using an overhead camera for feedback.
   • Accomplished the task of drawing shapes with maximum accuracy and minimum time.
4. Soil Monitoring Bot

   • Built the SM Bot for deployment on an arena that abstracts an agriculture field.
   • Powered by an FPGA (Field Programmable Gate Array) that controls its sensors, actuators, wireless communication, and pick-place mechanism.
   • The robot traverses the arena, senses the environment, and picks and places necessary supplies from the supply unit to the field.