Computational Models for Auditory Scene Analysis (2015 - 2019)

Auditory Attention via Gated Recurrent Neural Networks (2018 - 2019)

Theoretical models of how top-down attention could be used to facilitate auditory attention.

Real-Time Speech Enhancement on Android (May 2018 - August 2018)

Working with the Machine Perception team at Google we experimented with running deep learning models in real-time on android phones using a new framework, TfLite.

Generalization Challenges for Neural Networks Architectures​ (2017 - 2018)

In this work we studied the generalization performance of different neural network architectures on the task of audio source separation.

Discrete Music Generation with GANs (May 2017 - August 2017)

Working with the Magenta team at Google Brain we built Generative Adversarial Networks (GANs) for improving the quality of our RNN-based generative models of music.

Unsupervised learning of auditory stimuli (2017)

Working with Profession Olshausen we studied how the principles of efficient coding could be used to explain the spiked-based cochlear response of mammals.

​Voice Conversion (2015 - 2016)​

Working with Professor Bruna we applied convolutional neural networks and generative adversarial networks to convert the voices of audio samples.

Modeling the Syrinx of Birds for Call Synthesis (Dec 2015 - Feb 2016)

Working with Professor Theunissen we applied machine learning algorithms to syrinx models in order to better understand how zebra finches produce their calls.

Learning Transformational Invariants (2014 - 2015)

Working with Professor Olshausen we studied the dynamics of sparse representations of images using recurrent neural networks.

​Information-based learning by agents in unbounded state spaces (2014)

Working with Professor Sommer we built non-parametric bayesian models to study how animals might explore unknown and unbounded environments.