Anton Jeran Ratnarajah

Mail, Linkedin, GitHub, YouTube, Google Scholar, Researchgate, CV

I am a 5th PhD student in Electrical and Computer Engineering at University of Maryland-College Park advised by Prof. Dinesh Manocha. I am broadly interested in building machine learning models to create high quality sound experience in interactive applications. My current line of research is learning-based sound synthesis and propogation for AR and VR applications and audio/speech processing applications.

Research

	M3-AUDIODEC: MULTI-CHANNEL MULTI-SPEAKER MULTI-SPATIAL AUDIO CODEC Anton Jeran Ratnarajah, Shi-Xiong Zhang, Dong Yu Arxiv Code \| Paper \| Webpage
	Listen2Scene: Interactive material-aware binaural sound propagation for reconstructed 3D scenes Anton Jeran Ratnarajah, Dinesh Manocha Arxiv Code \| Paper \| Webpage
	Towards Improved Room Impulse Response Estimation for Speech Recognition Anton Jeran Ratnarajah, Ishwarya Ananthabhotla, Vamsi Krishna Ithapu, Pablo Hoffmann, Dinesh Manocha, Paul Calamia ICASSP 2023 Paper \| Webpage \| Video
	MESH2IR: Neural Acoustic Impulse Response Generator for Complex 3D Scenes Anton Jeran Ratnarajah, Zhenyu Tang,Rohith Aralikatti, Dinesh Manocha ACM Multimedia 2022 (Oral) Code \| Paper \| Video \| Webpage
	FAST-RIR: FAST NEURAL DIFFUSE ROOM IMPULSE RESPONSE GENERATOR Anton Jeran Ratnarajah, Shi-Xiong Zhang, Meng Yu, Zhenyu Tang, Dinesh Manocha, Dong Yu ICASSP 2022 Code \| Paper \| Video \| Webpage
	IR-GAN: Room impulse response generator for far-field speech recognition Anton Jeran Ratnarajah, Zhenyu Tang, Dinesh Manocha INTERSPEECH 2021 Code \| Video \| Paper \| Slides
	TS-RIR: Translated synthetic room impulse responses for speech augmentation Anton Jeran Ratnarajah, Zhenyu Tang, Dinesh Manocha Accepted to IEEE ASRU 2021 Code \| Video \| Paper \| Slides
	IMPROVING REVERBERANT SPEECH SEPARATION WITH MULTI-STAGE TRAINING AND CURRICULUM LEARNING Rohith Aralikatti, Anton Jeran Ratnarajah, Zhenyu Tang, Dinesh Manocha Accepted to IEEE ASRU 2021 Paper
	Moving Object Based Collision-Free Video Synopsis Anton Jeran Ratnarajah, Sahani Goonetilleke, Dumindu Tissera, Kapilan Balagopalan, Ranga Rodrigo 2018 IEEE International Conference on Systems, Man, and Cybernetics Paper

Education

	University of Maryland - College Park, College Park, MD, USA PhD in Electrical and Computer Engineering Cumulative GPA: 3.86 / 4 Aug. 2019 - Present (2024 expected)
	University of Moratuwa, Moratuwa, Sri Lanka Bachelor of Science in Engineering Cumulative GPA: 3.91 / 4.2 Feb. 2014 - Jan. 2018

Work Experience

	Research Intern Tencent America, Bellevue, Washington, United States May. 2023 - Aug. 2023 Implemented an innovative neural spatial audio codec for efficient compression of multi-channel (binaural) speech in both single and multi-speaker scenarios while retaining the spatial location information of each speaker.
	Research Scientist Intern META Reality Labs, Redmond, Washington, United States May. 2022 - Nov. 2022 Implemented a blind room impulse response estimation system in the context of far-field speech recognition.
	Research Intern Tencent America, Bellevue, Washington, United States May. 2021 - Aug. 2021 Implemented a neural-network-based fast diffuse room impulse response generator (FAST-RIR) for generating room impulse responses (RIRs) for a given acoustic environment.
	Engineer Wave Computing, Colombo, Sri Lanka Feb. 2018 - Jul. 2019 Developed machine learning applications for Wave Computing’s Dataflow Processing Unit (DPU) Architecture using Wave Flow Graph (WFG), a data flow description language developed by Wave Computing. Compiled and simulated the designs in Wave Computing’s complete EDA toolchain. Debugged and proposed suggestions to improve the toolchain and the architecture.
	Research Intern HESL Lab, Nanyang Technological University, Singapore Aug. 2016 - Dec. 2016 Successfully completed a project titled “Low complexity techniques for Vehicle Localization and Tracking" under the guidance of Dr. Lam Siew Kei .

Teaching Experience

ENEE 245: Digital Circuits and Systems Laboratory
Teaching Assistant for Professor Manoj Franklin
University of Maryland - College Park, College Park, MD, USA
Spring 2022

CMSC 742: Algorithms in Machine Learning: Guarantees and Analyses
Teaching Assistant for Professor Furong Huang
University of Maryland - College Park, College Park, MD, USA
Fall 2021

ENEE 630: Advanced Digital Signal Processing
Teaching Assistant for Professor K. J. Ray Liu
University of Maryland - College Park, College Park, MD, USA
Fall 2020

ENEE 425: Digital Signal Processing
Teaching Assistant for Professor Behtash Babadi
University of Maryland - College Park, College Park, MD, USA
Spring 2020

ENEE 425: Digital Signal Processing
Teaching Assistant for Professor Carol Espy-Wilson
University of Maryland - College Park, College Park, MD, USA
Fall 2019

	M3-AUDIODEC: MULTI-CHANNEL MULTI-SPEAKER MULTI-SPATIAL AUDIO CODEC Anton Jeran Ratnarajah, Shi-Xiong Zhang, Dong Yu Arxiv Code \| Paper \| Webpage
	Listen2Scene: Interactive material-aware binaural sound propagation for reconstructed 3D scenes Anton Jeran Ratnarajah, Dinesh Manocha Arxiv Code \| Paper \| Webpage
	Towards Improved Room Impulse Response Estimation for Speech Recognition Anton Jeran Ratnarajah, Ishwarya Ananthabhotla, Vamsi Krishna Ithapu, Pablo Hoffmann, Dinesh Manocha, Paul Calamia ICASSP 2023 Paper \| Webpage \| Video
	MESH2IR: Neural Acoustic Impulse Response Generator for Complex 3D Scenes Anton Jeran Ratnarajah, Zhenyu Tang,Rohith Aralikatti, Dinesh Manocha ACM Multimedia 2022 (Oral) Code \| Paper \| Video \| Webpage
	FAST-RIR: FAST NEURAL DIFFUSE ROOM IMPULSE RESPONSE GENERATOR Anton Jeran Ratnarajah, Shi-Xiong Zhang, Meng Yu, Zhenyu Tang, Dinesh Manocha, Dong Yu ICASSP 2022 Code \| Paper \| Video \| Webpage
	IR-GAN: Room impulse response generator for far-field speech recognition Anton Jeran Ratnarajah, Zhenyu Tang, Dinesh Manocha INTERSPEECH 2021 Code \| Video \| Paper \| Slides
	TS-RIR: Translated synthetic room impulse responses for speech augmentation Anton Jeran Ratnarajah, Zhenyu Tang, Dinesh Manocha Accepted to IEEE ASRU 2021 Code \| Video \| Paper \| Slides
	IMPROVING REVERBERANT SPEECH SEPARATION WITH MULTI-STAGE TRAINING AND CURRICULUM LEARNING Rohith Aralikatti, Anton Jeran Ratnarajah, Zhenyu Tang, Dinesh Manocha Accepted to IEEE ASRU 2021 Paper
	Moving Object Based Collision-Free Video Synopsis Anton Jeran Ratnarajah, Sahani Goonetilleke, Dumindu Tissera, Kapilan Balagopalan, Ranga Rodrigo 2018 IEEE International Conference on Systems, Man, and Cybernetics Paper