Ethan's head

Ethan Manilow is a final year PhD Candidate studying Computer Science working under Bryan Pardo in the Interactive Audio Lab at Northwestern University. He currently works part time as a Student Researcher at Google Brain on the Magenta Team after a research internship with them in the summer of 2020. Prior to that, he was a research intern at MERL on the Speech and Audio Team.

He currently lives in Chicago, where he spends his free time fingerpicking his acoustic guitar and smiling at dogs he passes on the sidewalk.


>>> Projects
  • Bespoke Score-Informed Nets 2020
    Purpose-built neural networks that separate sources in a mix by synthesizing their own training data. [demo]
  • audioLIME 2020
    Listenable explanations of black-box audio models using sources as interpretable components. (With Verena Haunschmid) [code]
  • Cerberus 2020
    Neural network to isolate instruments in a musical mix and simultaneously transcribe the notes of each instrument. [demo]
  • Slakh2100 2019
    Dataset for music source separation with 145 hours of mixtures, isolated sources, and MIDI. [website]
  • WHAM! 2019
    Dataset of nonstationary background noise for speech separation. [website]
  • SDR Prediction 2017
    Predicting source separation evaluation metrics without ground truth. [demo]
  • nussl 2015 - Now
    A python audio source separation library with implementation of many popular algorithms. [github] [docs]
  • SampSyn 2012
    [github]
>>> Publications
>>> CV [pdf]

Sometimes he goes on Twitter. His email address is his first and last name at Google's email service.