Yossi Adi
Chief Research Officer, Co-Founder at The Hebrew University of Jerusalem
Tel-Aviv District, Israel
Summary
🤩
Rockstar🎓
Top SchoolYossi Adi is an Assistant Professor and experienced researcher specializing in machine learning, deep learning, and speech/audio processing with over a decade of applied and academic experience. He holds a Ph.D. in Computer Science from Bar-Ilan University and has held research roles at Facebook AI and IBM, bridging cutting-edge research with practical engineering. Yossi has a strong publication record in top-tier ML and signal-processing venues and has contributed to prominent open-source projects such as Facebook Research’s svoice, improving data generation and audio handling for speaker separation. Based in the Tel-Aviv District, he combines academic rigor with hands-on model and dataset engineering, and brings a creative edge as a musician and founder of Lucille Crew.
11 years of coding experience
11 years of employment as a software developer
Bachelor of Science (B.Sc.), Computer Science, Bachelor of Science (B.Sc.), Computer Science at The College of Management Academic StudiesDoctor of Philosophy (Ph.D.), Computer Science, Doctor of Philosophy (Ph.D.), Computer Science at Bar-Ilan University
Github Skills (11)
data-generation10
pytorch10
machine-learning10
audio-processing10
python10
data-set9
datasets9
argument-parsing9
sound-files9
scipy8
numpy8
Programming languages (6)
Github contributions (5)
facebookresearch/svoice
Dec 2020 - Dec 2021
We provide a PyTorch implementation of the paper Voice Separation with an Unknown Number of Multiple Speakers In which, we present a new method for separating a mixed audio sequence, in which multiple voices speak simultaneously. The new method employs gated neural networks that are trained to separate the voices at multiple processing steps, while maintaining the speaker in each output channel fixed. A different model is trained for every number of possible speakers, and the model with the largest number of speakers is employed to select the actual number of speakers in a given sample. Our method greatly outperforms the current state of the art, which, as we show, is not competitive for more than two speakers.
Role in this project:
ML Engineer
ML EngineerContributions:1 review, 15 commits, 7 PRs in 11 months
Contributions summary:Yossi focused on data generation and model support within the speech separation project. They added scripts for generating datasets, including features for room simulation and noise addition. Furthermore, the user fixed issues related to argument parsing and audio loading, which improved the usability of the models. These modifications enhanced the development of the core model and its ability to process varied audio inputs.
audioneural-networkpytorch
aveni/WatermarkNN
Oct 2018 - Oct 2018
Watermarking Deep Neural Networks (USENIX 2018)
Contributions:19 commits in 1 day
deep-neural-networks
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.