Michael Auli
RESEARCH SCIENTIST | MENLO PARK, UNITED STATES
Michael is a Research Scientist at Facebook AI Research in Menlo Park. During his Ph.D., he worked on CCG parsing at the University of Edinburgh. While at Microsoft Research, Michael was involved in the early work on neural machine translation and neural dialogue models. Following this, he led the team that developed convolutional sequence to sequence models. Michael currently works on semi-supervised and self-supervised learning applied to natural language processing and speech recognition.
Michael's Publications
August 01, 2024
SPEECH & AUDIO
NLP
Toward Joint Language Modeling for Speech Units and Text
Ju-Chieh Chou, Karen Livescu, Alexis Conneau, Alexei Baevski, Wei-Ning Hsu, Arun Babu, Michael Auli
August 01, 2024
October 22, 2023
SPEECH & AUDIO
DinoSR: Self-Distillation and Online Clustering for Self-supervised Speech Representation Learning
Wei-Ning Hsu, Michael Auli, Alexander Liu, Heng-Jui Chang, James Glass
October 22, 2023
July 23, 2023
NLP
COMPUTER VISION
Efficient Self-supervised Learning with Contextualized Target Representations for Vision, Speech and Language
Michael Auli, Alexei Baevski, Arun Babu, Wei-Ning Hsu
July 23, 2023
May 22, 2023
SPEECH & AUDIO
NLP
Scaling Speech Technology to 1,000+ Languages
Sayani Kundu, Alexei Baevski, Alexis Conneau, Michael Auli, Ali Elkahky, Andros Tjandra, Apoorv Vyas, Arun Babu, Bowen Shi, Maryam Fazel-Zarandi, Paden Tomasello, Vineel Pratap, Wei-Ning Hsu
May 22, 2023
July 17, 2022
data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language
Michael Auli, Alexei Baevski, Arun Babu, Jiatao Gu, Wei-Ning Hsu, Qiantong Xu
July 17, 2022
October 25, 2021
October 25, 2021
December 06, 2020
wav2vec 2.0: A Framework for Self-SupervisedLearning of Speech Representations
Michael Auli, Abdelrahman Mohamed, Alexei Baevski, Henry Zhou
December 06, 2020
November 18, 2020
RESEARCH
NLP
Language Models not just for Pre-training: Fast Online Neural Noisy Channel Modeling
Michael Auli, Sergey Edunov, Shruti Bhosale, Kyra Yee
November 18, 2020
April 16, 2020
RESEARCH
NLP
vq-wav2vec: Self-Supervised Learning of Discrete Speech Representations
Alexei Baevski, Michael Auli, Steffen Schneider
April 16, 2020
October 23, 2019
RESEARCH
SPEECH & AUDIO
Cloze-driven Pretraining of Self-attention Networks
Michael Auli, Alexei Baevski, Luke Zettlemoyer, Sergey Edunov, Yinhan Liu
October 23, 2019
August 29, 2019
RESEARCH
SPEECH & AUDIO
wav2vec: Unsupervised Pre-training for Speech Recognition
Michael Auli, Alexei Baevski, Ronan Collobert, Steffen Schneider
August 29, 2019
August 27, 2019
RESEARCH
Simple and Effective Noisy Channel Modeling for Neural Machine Translation
Michael Auli, Kyra Yee, Yann N. Dauphin
August 27, 2019
June 08, 2019
RESEARCH
NLP
Mixture Models for Diverse Machine Translation: Tricks of the Trade
Marc'Aurelio Ranzato, Michael Auli, Myle Ott, Tianxiao Shen
June 08, 2019
May 27, 2019
RESEARCH
SPEECH & AUDIO
fairseq: A Fast, Extensible Toolkit for Sequence Modeling
Michael Auli, Alexei Baevski, Angela Fan, Myle Ott, Nathan Ng, Sam Gross, Sergey Edunov, David Grangier
May 27, 2019
May 13, 2019
RESEARCH
COMPUTER VISION
3D human pose estimation in video with temporal convolutions and semi-supervised training
Michael Auli, Christoph Feichtenhofer, Dario Pavllo, David Grangier
May 13, 2019
April 19, 2019
RESEARCH
SPEECH & AUDIO
Pay less attention with Lightweight and Dynamic Convolutions
Michael Auli, Alexei Baevski, Angela Fan, Felix Wu, Yann Dauphin
April 19, 2019
October 31, 2018
RESEARCH
SPEECH & AUDIO
Understanding Back-Translation at Scale
Michael Auli, Myle Ott, Sergey Edunov, David Grangier
October 31, 2018
October 30, 2018
October 30, 2018
September 03, 2018
RESEARCH
SPEECH & AUDIO
QuaterNet: A Quaternion-based Recurrent Model for Human Motion
David Grangier, Dario Pavllo, Michael Auli
September 03, 2018
July 13, 2018
RESEARCH
SPEECH & AUDIO
Analyzing Uncertainty in Neural Machine Translation
Myle Ott, David Grangier, Marc'Aurelio Ranzato, Michael Auli
July 13, 2018
Our approach
Latest news
Foundational models
Meta © 2026