AI People Wiki

Builders & open source

Yonghui Wu

First author of the paper on Google's Neural Machine Translation system (GNMT); co-author of Lingvo

Yonghui Wu is the first author of the 2016 paper presenting GNMT, Google's Neural Machine Translation system, which describes deep LSTM encoder and decoder networks with attention for translation and thanks the Google Brain and Google Translate teams. Wu is also a co-author of the 2019 paper on Lingvo, a TensorFlow framework for sequence-to-sequence modeling published in the TensorFlow GitHub organization under the Apache 2.0 license.123

At a glance

Current role
None confirmed in the sources
Work listed
2
Sources cited
7
Profile reviewed

Research

Wu is first author of the 2016 paper describing GNMT, Google's Neural Machine Translation system, which used deep LSTM encoder and decoder networks with attention and acknowledges the Google Brain and Google Translate teams. Wu also co-authored the 2019 paper on Lingvo, a TensorFlow framework for sequence-to-sequence modeling whose code is published in the TensorFlow GitHub organization.123

Wu's co-authored work extends across speech and vision. In speech, it includes Tacotron 2, a 2017 architecture for synthesizing speech directly from text, and Conformer, a 2020 model combining convolution with Transformers for speech recognition. In 2022 Wu co-authored CoCa, a contrastive captioning approach to image-text foundation models, and Parti, an autoregressive text-to-image model.4567

Work and contributions

  • Google's Neural Machine Translation system (GNMT) Published research

    First author of the 2016 paper presenting GNMT, a translation model built from a deep LSTM network with eight encoder and eight decoder layers and an attention mechanism, using sub-word units called wordpieces.1

  • Lingvo Open-source code

    Co-author of the 2019 Lingvo paper; Lingvo is a framework for building neural networks in TensorFlow, particularly sequence models, released under the Apache 2.0 license.23

Timeline

  1. Released the GNMT paper on Google's Neural Machine Translation system as first author.1
  2. Co-authored Tacotron 2, a neural architecture for speech synthesis from text.4
  3. Co-authored the Lingvo paper on a framework for sequence-to-sequence modeling.2
  4. Co-authored Conformer, a convolution-augmented Transformer for speech recognition.5
  5. Co-authored CoCa, contrastive captioners as image-text foundation models.6
  6. Co-authored Parti, an autoregressive model for content-rich text-to-image generation.7

Documented roles

The sources do not confirm a current role; these are the roles they document.

  • First author of Google's 2016 GNMT paper, Google1

Companies in their history

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

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

  1. Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation (arXiv 1609.08144)arXiv · Paper · published Sep 26, 2016 · accessed Oct 21, 2026
  2. Lingvo: a Modular and Scalable Framework for Sequence-to-Sequence Modeling (arXiv 1902.08295)arXiv · Paper · published Feb 21, 2019 · accessed Oct 21, 2026
  3. tensorflow/lingvo README (Apache-2.0)TensorFlow (GitHub repository) · Repository · accessed Oct 21, 2026
  4. Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram PredictionsarXiv · Paper · published Dec 16, 2017 · accessed Oct 21, 2026
  5. Conformer: Convolution-augmented Transformer for Speech RecognitionarXiv · Paper · published May 16, 2020 · accessed Oct 21, 2026
  6. CoCa: Contrastive Captioners are Image-Text Foundation ModelsarXiv · Paper · published May 4, 2022 · accessed Oct 21, 2026
  7. Scaling Autoregressive Models for Content-Rich Text-to-Image GenerationarXiv · Paper · published Jun 22, 2022 · accessed Oct 21, 2026