AI People Wiki

Builders & open source

Yu-An Chung

Research Scientist, Facebook AI Research (FAIR), Meta

Yu-An Chung is a research scientist at Facebook AI Research (FAIR), with research interests in representation learning, self-supervised learning, and generative modeling for speech, and a current focus on real-time direct speech-to-speech translation. Chung completed a PhD in computer science at MIT under James Glass in 2022. Chung is listed in the technical leadership and project management group of Omnilingual ASR, FAIR at Meta's 2025 open-source speech recognition system covering more than 1,600 languages, and appears with FAIR at Meta on the 2026 Omnilingual SONAR paper on cross-lingual and cross-modal sentence embeddings.123

At a glance

Current role
Research Scientist, Facebook AI Research (FAIR)
Company
Meta
Sector
Company
Work listed
2
Sources cited
8
Profile reviewed

Career

Yu-An Chung earned an SM in 2019 and a PhD in 2022 in computer science at MIT under the supervision of James Glass. During the PhD, Chung interned on speech synthesis with Google Perception, on text summarization and on speech recognition and translation at Google Brain, and on spoken language understanding at Microsoft Cognitive Services Research.1

Chung is now a research scientist at Facebook AI Research, focusing on real-time direct speech-to-speech translation.1

Research

Many of Chung's papers from the MIT years study self-supervised speech representations. With Glass, Chung introduced Speech2Vec (2018), which learns word embeddings from speech with a sequence-to-sequence model; with co-authors including Glass, Chung also proposed an unsupervised autoregressive model for speech representation learning (2019), followed by further papers on autoregressive predictive coding.456

Chung was first author of w2v-BERT (2021), which combines contrastive learning and masked language modeling for speech pre-training. At FAIR, Chung is listed in the technical leadership of Omnilingual ASR (2025) and co-authored Omnilingual SONAR (2026), on sentence embeddings shared across text and speech in many languages.723

Work and contributions

  • Omnilingual ASR Open-source code

    Listed under technical leadership and project management on the 2025 Omnilingual ASR paper from FAIR at Meta; the repository is released under the Apache 2.0 license.28

  • Omnilingual SONAR Published research

    Co-author, listed with FAIR at Meta, of the 2026 paper on cross-lingual and cross-modal sentence embeddings spanning massively multilingual text and speech.3

Timeline

  1. Released Speech2Vec, a sequence-to-sequence framework for word embeddings from speech.4
  2. Released An Unsupervised Autoregressive Model for Speech Representation Learning.5
  3. Released w2v-BERT for self-supervised speech pre-training.7
  4. Completed a PhD in computer science at MIT under James Glass.1
  5. Listed in the technical leadership of the Omnilingual ASR paper from FAIR at Meta.2
  6. Co-authored the Omnilingual SONAR paper on cross-lingual and cross-modal sentence embeddings.3

Where they work

  • Meta 106 people in the wiki

Suggest a correction or removal

Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.

References

8 sources · profile reviewed Oct 21, 2026 · data updated Oct 21, 2026

  1. Yu-An Chung — homepageYu-An Chung · Official page · accessed Oct 21, 2026
  2. Omnilingual ASR: Open-Source Multilingual Speech Recognition for 1600+ Languages (arXiv 2511.09690)arXiv · Paper · published Nov 12, 2025 · accessed Oct 21, 2026
  3. Omnilingual SONAR: Cross-Lingual and Cross-Modal Sentence Embeddings Bridging Massively Multilingual Text and Speech (arXiv 2603.16606)arXiv · Paper · published Mar 17, 2026 · accessed Oct 21, 2026
  4. Speech2Vec: A Sequence-to-Sequence Framework for Learning Word Embeddings from Speech (arXiv 1803.08976)arXiv · Paper · published Mar 23, 2018 · accessed Oct 21, 2026
  5. An Unsupervised Autoregressive Model for Speech Representation Learning (arXiv 1904.03240)arXiv · Paper · published Apr 5, 2019 · accessed Oct 21, 2026
  6. Yu-An Chung — papersYu-An Chung · Official page · accessed Oct 21, 2026
  7. W2v-BERT: Combining Contrastive Learning and Masked Language Modeling for Self-Supervised Speech Pre-Training (arXiv 2108.06209)arXiv · Paper · published Aug 7, 2021 · accessed Oct 21, 2026
  8. facebookresearch/omnilingual-asr LICENSE (Apache 2.0)facebookresearch (GitHub repository) · License · accessed Oct 21, 2026