Atlas / Reports / Detail
MiniMax-Speech: Intrinsic Zero-Shot Speech Understanding for Advanced Foundation Models
Speech Language Models report from MiniMax with 10 connected researchers in the LLMpeople atlas.
Connected researchers
Yang Yue
MiniMax / Moonshot AI
Researcher at Moonshot AI and co-author of the Kimi K2.5 report on visual agentic intelligence.
Yusheng Zhao
MiniMax
Research scientist at MiniMax AI Research focused on reinforcement learning, reasoning, multimodal learning, large language models, and large-scale distributed systems. He received a PhD in machine learning from Carnegie Mellon University.
Xiang Li
MiniMax
Xiang Li is listed as an author of the MiniMax technical report MiniMax-01: Scaling Foundation Models with Lightning Attention.
Qingyang Ge
MiniMax
Public report authorship links Qingyang Ge to the MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention at MiniMax.
Dawei Feng
MiniMax
Co-founder and research scientist at MiniMax AI Research. He received a PhD from Tsinghua University and works on foundation models, reinforcement learning, and data systems, with publications at major machine learning and NLP venues.
Wenchao Zhou
MiniMax
Public profiles describe Wenchao Zhou as Director of Data Product and Data Analytics at Alibaba Cloud Intelligence and a former tenured computer science faculty member at Georgetown University. His work centers on databases and distributed systems.
Jinyuan Jia
MiniMax
Researcher working on speech and multimodal language models, including MiniMax-Speech and related speech understanding work.
Ming Ding
MiniMax
Lead of foundation models at MiniMax working on large language models, multimodal pretraining, and efficient training systems. He completed a PhD in computer science at Tsinghua University.
Dingchen Yang
MiniMax
Dingchen Yang is listed as an author of the MiniMax technical report MiniMax-Speech: Intrinsic Zero-Shot Speech Understanding for Advanced Foundation Models.
Huan Chen
MiniMax
Researcher at MiniMax and coauthor of the MiniMax-Speech: Intrinsic Zero-Shot Speech Understanding for Advanced Foundation Models.