Ana Elisa Mendez Mendez

Ana Elisa Méndez Méndez

I am a Ph.D. in Music Technology recently graduated from the Music and Audio Research Lab at New York University, advised by Prof. Juan Pablo Bello. I previously obtained a Master's degree in Music Technology from NYU, and got my Bachelor's degree in Systems Engineering from Universidad Metropolitana in Venezuela.

I successfully defended my dissertation in the fall of 2023. In the past years, I have worked as a research assistant for two NSF-funded projects: Sounds of New York City (SONYC) and S3D: Spatial Sound Scene Descripition, focusing on data annotation for solving machine learning problems such as sound event detection and sound event localization. More specifically, I've worked on understanding how to collect human annotations and which samples should be labeled through active learning. I also spent the summers of 2019, 2020, and 2021 as a research intern at Microsoft Corporation for the BAG organization working on data annotation techniques, active learning, and tools for ML expert audio annotation. Finally, I spent the summer of 2023 as a deep learning research intern for Bosch, working on active learning techniques through reinforcement learning.

Email / CV / Google Scholar / LinkedIn /

Research

I'm interested in the intersection of machine learning and human-computer interaction, focusing data annotation processes of audio datasets for which "ground truth" labels are unknown. I am especially interested in creating an annotation framework that enables researchers in machine listening to generate large-scale datasets at a substantially reduced annotation effort while maintaining high quality.

Publications

Conference Papers

	Eliciting Confidence for Improving Crowdsourced Audio Annotations PDF Ana Elisa Méndez Méndez, Mark Cartwright, Juan Pablo Bello, Oded Nov In Proceedings of the ACM on Human-Computer Interaction, Volume 6, Issue CSCW1, 2022
	SONYC-UST-V2: An Urban Sound Tagging Dataset with Spatiotemporal Context PDF Mark Cartwright, Jason Cramer Ana Elisa Méndez Méndez, et. al In Detection and Classification of Acoustic Scenes and Events (DCASE), 2020
	Multi-label Sound Event Retrieval Using a Deep Learning-based Siamese Structure with a Pairwise Presence Matrix PDF Jianyu Fan, Eric Nichols, Daniel Tompkins, Ana Elisa Méndez Méndez, Benjamin Elizalde, Philippe Pasquier In 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020
	SONYC Urban Sound Tagging (SONYC-UST): A Multilabel Dataset from an Urban Acoustic Sensor Network PDF Mark Cartwright, Ana Elisa Méndez Méndez, Jason Cramer, et. al In Detection and Classification of Acoustic Scenes and Events (DCASE), 2019
	Active Learning for Efficient Audio Annotation and Classification with a Large Amount of Unlabeled Data PDF Yu Wang, Ana Elisa Méndez Méndez, Mark Cartwright, Juan Pablo Bello In Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2019
	Machine-Crowd-Expert model for increasing user engagement and annotation quality PDF / Poster Ana Elisa Méndez Méndez, Mark Cartwright, Juan Pablo Bello In Proceedings of ACM CHI Conference Extended Abstracts on Human Factors in Computing Systems (CHI EA), 2019
	Crowdsourcing multi-label audio annotation tasks with citizen scientists PDF Mark Cartwright, Graham Dove, Ana Elisa Méndez Méndez, Juan Pablo Bello, Oded Nov In Proceedings of ACM Conference on Human Factors in Computing Systems (CHI), 2019
	CityTones: A Repository of Crowdsourced Annotated Soundfield Soundscapes PDF Agnieszka Roginska, Hyunkook Lee, Ana Elisa Méndez Méndez, Scott Murakami, Andrea Genovese In Audio Engineering Society Convention 146, 2019