Technical Projects
Here are the projects I've completed during my academic studies and as part of my work experience.
Spectral Enhancement Neural Audio Processor (GenAI for Audio)
Software Developer and Engineer
The solution employs a novel hybrid approach that combines traditional signal processing with neural networks. It features innovative frequency-domain processing, multi-resolution analysis with multiple FFT sizes and cross-band attention, and advanced phase reconstruction, all optimized through specialized loss functions such as MEL, STFT, and adversarial losses.
Live Music Assessment Website (React + Flask)
Software Developer and Engineer
Developed frontend and backend solution for live music assessment web application using React framework, Web Audio API, MusicXML, Flask framework, Python along with DSP Processing along with API development.
​
This project is one of its kind in the country, and its a big achievement to do the task real-time on the web interface.
Live MIDI Assessment Application (ElectronJS + Python)
Software Developer/ Software Engineer
Created a desktop application and scoring algorithms for Windows using ElectronJS and Python for real-time MIDI evaluation in classrooms with several MIDI Controllers.
​
The application is live and used for regular examinations in school across Mumbai.
.png)
Music Genre Classification using Deep Neural Networks (Python)
Capstone Project (AI and ML - IIT Guwahati)
Sound/Audio signals can be represented in the form of various parameters such as frequency, bandwidth, roll-off and so on. Using various python libraries, performed a feature extraction for these audio signals. These features can then be processed and further used to perform classification.
​
Dataset Used - GTZAN Dataset
Features - Mel-Frequency, Cepstral Coefficients, Spectral Centroid, Zero Crossing Rate, Chroma Frequencies and Spectral Roll-off.
Libraries - Numpy, Pandas, Scipy, Librosa, Sklearn, Matplotlib, Tensorflow, Keras.
Implementation of Embedded Machine Learning for Medical Assistance by Robotic Systems (React + Flask + Python + SQL + C++ )
MTech - Research Project
The goal of the project is to create a consumer ready IoT Framework for Medical Technology.
​
The project has AI enabled VUI, Live Chat functionality, Live Patient Vital monitoring and also a Robot Interface which can be accessed through a web application.
Simulation and Design of a low power Integrated Processor. (VHDL - VLSI)
BTech - Research Project
An Arithmetic Logic Unit which is 16 bits wide is capable of performing all logical operations as well as all arithmetic operations allowing low voltage operation.
​
Softwares Used: Aldec Active HDL, ISE Simulator
