profile photo

Sritharan Braveenan

Official Email: Sritharan[DOT]Braveenan[AT]uga[DOT]edu

Personal Email: braveenans[AT]gmail[DOT]com

Other Email: braveenans[DOT]22[AT]cse[DOT]mrt[DOT]ac[DOT]lk, braveenans[DOT]22[AT]uom[DOT]lk

I am currently pursuing a Doctor of Philosophy (PhD) in Computer Science at the School of Computing, University of Georgia. I hold a research-based master's degree in Computer Science and Engineering from the University of Moratuwa, Sri Lanka, where my research focused on speech processing. My academic journey began with a bachelor's degree in Electronic and Telecommunication Engineering, also from the University of Moratuwa.

During my undergraduate studies, I gained professional experience through internships at Wave Computing and Paraqum Technologies. Before starting my PhD, I worked as a Software Engineer at Axiata Digital Labs, further honing my skills in software development and engineering.

My previous research includes the development of a unified voice model for speech, speaker, and emotion recognition using multi-task learning, as well as contributions to light field video compression algorithms utilizing approximate discrete cosine transformation. I also explored vision-based autonomous drones for GPS-denied environments during my undergraduate studies.

My research interests lie at the intersection of signal processing and machine learning, with a focus on speech processing, multimedia compression, and autonomous systems.

CV  |  LinkedIn  |  Google Scholar  |  GitHub

Recent Updates
  • [2025-12-21] Presented wavCSE: Learning Fixed-size Unified Speech Embeddings via Feature-based Multi-Task Learning at IJCNLP–AACL 2025, Mumbai, India.
  • [2025-03-18] Presented SUPERB-EP: Evaluating Encoder Pooling Techniques in Self-Supervised Learning Models for Speech Classification at the 2025 IEEE Symposium on Computational Intelligence in Natural Language Processing and Social Media (CI-NLPSoMe), IEEE.
  • [2025-01-20] Presented Advancing Multilingual Speaker Identification and Verification for Indo-Aryan and Dravidian Languages at the First Workshop on Natural Language Processing for Indo-Aryan and Dravidian Languages.
  • [2025-01-09] Published Computationally Efficient Light Field Video Compression Using 5-D Approximate DCT in the journal Journal of Low Power Electronics and Applications, MDPI.
  • [2025-01-01] Commenced my Doctor of Philosophy (PhD) program at the School of Computing at the University of Georgia.
  • [2024-08-31] Graduated with a Master of Science (Major Component Research) specialized in Computer Science and Engineering.
  • [2024-08-01] Joined as Lecturer on contract at the Department of CSE, University of Moratuwa.
  • More...
Publications
  1. wavCSE: Learning Fixed-size Unified Speech Embeddings via Feature-based Multi-Task Learning
    Sritharan, B., & Thayasivam, U.
    In Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics, 2025.
    Paper

  2. SUPERB-EP: Evaluating Encoder Pooling Techniques in Self-Supervised Learning Models for Speech Classification
    Sritharan, B., Thayasivam, U., & Bandara, S. J.
    In 2025 IEEE Symposium on Computational Intelligence in Natural Language Processing and Social Media (CI-NLPSoMe), IEEE, 2025.
    Paper

  3. Advancing Multilingual Speaker Identification and Verification for Indo-Aryan and Dravidian Languages
    Sritharan, B., & Thayasivam, U.
    In Proceedings of the First Workshop on Natural Language Processing for Indo-Aryan and Dravidian Languages, 2025.
    Paper

  4. Computationally Efficient Light Field Video Compression Using 5-D Approximate DCT
    Sritharan, B., Edussooriya, C. U., Wijenayake, C., Cintra, R. J., & Madanayake, A.
    Journal of Low Power Electronics and Applications, 15(1), 2, 2025.
    Paper
Current Research
  • Cross-View Geo-Localization (CVGL)
    Developing cross-view image matching methods for localizing ground-level imagery against geo-referenced aerial or satellite imagery in GPS-denied environments. Investigating viewpoint-invariant representation learning and robust camera pose estimation for vision-based navigation.
Other Research
  • Vision-Based Autonomous Inventory Management Drone
    Designed and implemented a vision-guided autonomous drone system for inventory monitoring and navigation in GPS-denied indoor environments. Video
Last Updated: February 12, 2026