Tan Xu
Xu Tan is a researcher specializing in multimodality and generative AI for video and audio, currently working on multimodal research at Tencent Hunyuan. He previously served as Research VP of Multimodality at Moonshot AI (Kimi) and, before that, as a Principal Research Manager at Microsoft Research Asia. He is best known for designing influential models and systems including FastSpeech, NaturalSpeech, MASS, MPNet, and HuggingGPT, with his research widely deployed in products such as Azure TTS, Microsoft Translator, and Kimi.
“The first person was from the Microsoft Research Asia side — Tan Xu. He'd also come from Microsoft. And LateTalk also broke the story of him joining Kimi before that.”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.