About

• Xu Tan is a Principal Research Manager at Machine Learning Group, Microsoft Research Asia (MSRA). His research interests cover generative AI and language/speech/music/avatar processing.
• He has published papers on top AI conferences, with 10000+ citations, and designed several popular language/speech/music models and systems, such as MASS, FastSpeech 1/2, NaturalSpeech 1/2, and Muzic.
• He has many technologies (e.g., neural machine translation; pre-training models: MASS; text to speech: FastSpeech 1/2, AdaSpeech, LRSpeech; AI music composition: TeleMelody, HiFiSinger; ASR error correction: FastCorrect 1/2; etc) deployed in Microsoft products (e.g., Bing Search/Ads, Microsoft Translator, Azure TTS, Azure ASR, Microsoft Xiaoice, etc).
• He and the team have several opensource projects on Github, such as MASS, MPNet (Huggingface), Muzic, NeuralSpeech, and HuggingGPT/JARVIS.
• He is an Action Editor of Transactions on Machine Learning Research (TMLR), an Area Chair or Meta Reviewer of AAAI/NeurIPS/ICASSP, a senior member of IEEE, and a member of the standing committee on Computational Art in China Computer Federation (CCF).

For a more detailed homepage, please visit my Microsoft Homepage.

We are hiring!

• We are hiring researchers on Multimodality, Machine Translation/NLP, Speech, Generative Models, and Deep Learning! If you are interested, welcome to contact me: tanxu2012@gmail.com.
• We are also hiring research interns in the above directions.