Portrait of Yan Huang

Yan Huang

Principal Applied Scientist


Yan’s research interests include machine learning and its applications in speech and natural language processing. Since she joined Microsoft in 2010, she has been working on improving acoustic model accuracy and robustness for Microsoft speech product. Her current research topics include large-scale semi-supervised deep learning acoustic model, improving acoustic model robustness with respect to phonetic and non-phonetic acoustic variability, model adaptation, and far-talk speech recognition.

Before joining Microsoft, she was with center of speech and language processing (CLSP) in Johns Hopkins University and International Computer Science Institute (ICSI) in UC Berkeley.