AI People Wiki

Builders & open source

Hsiao-Wuen Hon

Emeritus Researcher, Microsoft

  • Emeritus Researcher, Microsoft Research1
  • Independent board member, MediaTek1
  • Independent board member, Ambarella1
  • Vice-Chancellor Visiting Professor, The Chinese University of Hong Kong1

Hsiao-Wuen Hon is an Emeritus Researcher at Microsoft Research after a 28-year career at Microsoft, where Hon served as Corporate Vice President of the Asia-Pacific R&D Group and Managing Director of Microsoft Research Asia. Hon was a founder of the Microsoft Search Technology Center, which helped deliver the Bing search stack for Asia-Pacific, and earlier led Microsoft's Natural Interaction Service Division, which shipped Microsoft Speech Server. Hon's work spans speech recognition, natural language processing, and internet search, and Hon is a co-author of the 2019 UniLM paper on unified pre-training for language understanding and generation. Hon is an IEEE Fellow.12

At a glance

Current roles
Emeritus Researcher +3 more
Sector
Company
Education
Carnegie Mellon
Work listed
2
Honors
1
Sources cited
3
Profile reviewed

Career

Hsiao-Wuen Hon earned a PhD in computer science at Carnegie Mellon University under Raj Reddy, and earlier in Hon's career was Technology Director at Apple for the Chinese Dictation Kit. During a 28-year career at Microsoft, Hon led the Natural Interaction Service Division, which shipped Microsoft Speech Server and the Natural User Interface platform, and was a founder of the Microsoft Search Technology Center, which helped deliver the Bing search stack for Asia-Pacific.1

Hon served as Corporate Vice President of Microsoft's Asia-Pacific R&D Group and Managing Director of Microsoft Research Asia, and established the Microsoft Innovation Partnership Center for industry collaboration on AI. Hon is now an Emeritus Researcher at Microsoft Research, an independent board member of MediaTek and Ambarella, and a Vice-Chancellor Visiting Professor at The Chinese University of Hong Kong.1

Research

Hon's research spans speech recognition, natural language processing and internet search. With Kai-Fu Lee and Raj Reddy, Hon co-authored a 1990 overview of SPHINX, a large-vocabulary, speaker-independent continuous speech recognition system based on hidden Markov models, and in 2019 Hon co-authored UniLM, a Transformer pre-trained for both language understanding and generation.132

Work and contributions

  • UniLM (Unified Language Model Pre-training) Published research

    Co-author of the NeurIPS 2019 paper presenting UniLM, a pre-trained Transformer that can be fine-tuned for both language understanding and generation; the paper links its code and pre-trained models.2

  • Microsoft Speech Server and the Bing search stack for Asia-Pacific

    Led the Natural Interaction Service Division, which shipped Microsoft Speech Server and the Natural User Interface platform, and was a founder of the Microsoft Search Technology Center, which helped deliver the Bing search stack for Asia-Pacific.1

Timeline

  1. Co-authored "An overview of the SPHINX speech recognition system" with Kai-Fu Lee and Raj Reddy.3
  2. Co-authored UniLM, a unified pre-trained model for language understanding and generation.2

Other documented roles

Roles the sources document that are not confirmed as current.

  • Corporate Vice President, Asia-Pacific R&D Group, and Managing Director, Microsoft Research Asia, Microsoft1
  • Technology Director for the Chinese Dictation Kit, Apple1

Honors and recognition

  • IEEE Fellow · Institute of Electrical and Electronics Engineers1

Where they work

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

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

  1. Hsiao-Wuen Hon at Microsoft ResearchMicrosoft Research · Official page · accessed Oct 21, 2026
  2. Unified Language Model Pre-training for Natural Language Understanding and Generation (arXiv 1905.03197)arXiv · Paper · published May 8, 2019 · accessed Oct 21, 2026
  3. An overview of the SPHINX speech recognition systemCarnegie Mellon University Robotics Institute · Paper · accessed Oct 21, 2026