Skip to content
View shreyanc's full-sized avatar
  • Johannes Kepler University Linz
  • Linz, Austria

Block or report shreyanc

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please donโ€™t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this userโ€™s behavior. Learn more about reporting abuse.

Report abuse
shreyanc/README.md

Shreyan Chowdhury, Ph.D.

Machine Learning Engineer & Researcher โ€ข Audio / Music / Voice

Developing technologies that enhance the human listening and visual experience through the intersection of signal processing, machine learning, and creativity.


๐Ÿš€ Currently Building

๐ŸŽฎ Space Moths - Quantum AI-powered level generation for gaming
๐ŸŽผ RECURSE - Infinite music streaming powered by quantum-classical hybrid ML models
๐ŸŽ›๏ธ Neural Audio Processing - Real-time, low-latency models for professional audio applications


๐ŸŽฏ What I Do

๐ŸŽต Music Information Retrieval
๐Ÿ”Š Audio Signal Processing
๐Ÿค– Explainable AI
โšก Real-time ML Systems
๐ŸŽน Computational Musical Creativity

๐Ÿ› ๏ธ Technical Toolkit

Deep Learning & Audio

PyTorch TensorFlow scikit-learn

Production & MLOps

Docker AWS FastAPI

Languages

Python C++ MATLAB


๐Ÿ”ฌ Research Highlights

Explainable AI in Music ๐ŸŽผ

Making music emotion recognition interpretable and transferable

  • Developed novel approaches to decode emotional expression in piano performances
  • Created domain adaptation techniques for cross-dataset model transfer
  • Published in top-tier conferences (ICML, ICASSP, ISMIR)

Real-time Audio Processing โšก

From research to production-ready systems

  • Optimized neural models for sub-10ms latency audio processing
  • Built end-to-end condition monitoring systems deployed on embedded devices
  • Improved architectures for real-time guitar amplifier modeling

Quantum-Classical Hybrid ML ๐Ÿ”ฎ

Exploring the what quantum-classical hybrid ML systems can achieve in the creative space

  • Productionizing quantum reservoir learning for music generation and game level generation
  • Leading research at the intersection of quantum computing and creative AI

๐Ÿ“ซ Let's Connect

Pinned Loading

  1. model_debugging model_debugging Public

    Music emotion model debugging using midlevel feature and audioLIME explanations

    Jupyter Notebook 3 2

  2. midlevel_domain_adaptation midlevel_domain_adaptation Public

    Domain adaptation for midlevel features to piano audio (code for ICASSP 2021 paper)

    Python 1

  3. midlevel_general midlevel_general Public

    Training and inference of mid-level perceptual features from musical audio

    Python 4

  4. music-concept-bottleneck music-concept-bottleneck Public

    Applying concept bottleneck models for music analysis

    Python 2