B.S., Communication Engineering, Civil Aviation University of China, Tianjin, China
M.S., Information & Communication Engineering, Zhejiang University, Hangzhou, China
Ph.D, Electronic Engineering, Tsinghua University, Beijing, China
2011-2013, Postdoctoral fellow, Electronic Engineering, Tsinghua University, Beijing, China
2013-2018, Assistant Professor, Electronic Engineering, Tsinghua University, Beijing, China
2018-, Associate Professor, Electronic Engineering, Tsinghua University, Beijing, China
Speaker recognition, speaker diarization, speech recognition, audio event detection, language recognition, voice activity detection, speech enhancement and audio indexing.
B.S., Communication Engineering, Zhengzhou University, Henan, China
My work focus on audio topic classification. Using transformer and deep neural network to build an end-to-end system. I’m also interested in text classification and transformer, the former can help me implement audio topic classification, the latter parallelizes efficiently.
Audio topic classification, end-to-end, speech recognition, deep learning.
B.S., Optoelectronics Engineering, Tianjin University, Tianjin, China
My work mainly focus on speaker recognition. I make efforts to get a comprehensive understanding of this field through investigating related researches. Recently, I pay more attention to ASVspoof challenge and do in-depth study in this direction. As a postgraduate student, during this period, I do hope to make full use of learning resources to get self-improvement and then embrace the future in better shape.
Speaker recognition, spoof speech detection (antispoof), deep neural network.
M.S. candidate, Astronautical Engineering, Space Engineering University, Beijing, China
My main research direction is to identify the motion parameters of non-cooperative space targets such as unstable satellites and space debris by using the fusion algorithm of vision and optical vortex detection under the new system, so as to achieve the real-time processing ability under space-borne conditions.
Automatic speech recognition, computer vision.
B.S., Beijing Institute of Technology, Beijing, China
My research direction is the field of speaker diarization, and through continuous learning in this direction, I have a unique perspective on this field.
Speaker diarization, deep neural network
B.S., Department of Electronic Engineering , Tsinghua University, Beijing, China
I will start my research in the field of knowledge graphs. Knowledge graphs are important technologies in the fields of NLP and search engines. I hope that they can also be used in laboratory research on speech topic analysis.
Knowledge graphs, NLP.
B.S., Institute of Information Science and Management, Henan Agricultural University, Henan, China
My research direction is to estimate the importance of voice content, hoping that through hard work and research, the content importance estimate can be improved. I also hope to pay more attention to the development of the industry and make your own contribution in addition to the research.
Automatic speech recognition, natural language understanding.
B.S., XinJiang University, XinJiang, China
My research direction is the knowledge map direction of NLP. The main research content is: information extraction , knowledge mining and reasoning.
Information extraction, knowledge mining and reasoning.
B.S.,Communication Engineering,Tianshui Normal University.Gansu,China
My work direction is speaker recognition, mainly studying voice print + face multi-mode fusion recognition. Through reading the literature, I have a preliminary understanding of this field. I hope I can make further improvement through future study.
Speaker recongnition, voice print + face multi-mode fusion recognition, deep neural network.
B.S.,Network Engineering,Anyang Institute of Technology.Henan,China
My research direction is mainly the application of graph neural network in voiceprint recognition, and I am also very interested in knowledge graphs. Through reading the literature, I have a preliminary understanding of this field, and I hope that I can further improve through future learning.
Speaker recongnition, graph neural networks.
B.S., Textile Engineering, Tianjin Polytechnic University, Tianjin, China
My research direction is in the field of voiceprint recognition.I hope to improve myself by constantly learning domain knowledge and mastering relevant technologies.I hope it can also provide help for the related research in the laboratory.
Speaker recognition,automatic speech recognition,deep learning.
B.M., Information management and information system, Business College of Southwest University, Chongqing, China
I am interested in audio event detection, research on audio event detection and scene analysis, audio event detection of scarce data, weak supervised learning and unsupervised audio event detection.
audio event detection, deep learning.
B.E., College of Electronic Information Engineering, Beihang University, Beijing, China
My research direction is audio event detection, mainly through machine learning, especially deep learning method, to analyze and process audio data. Under unsupervised condition, detect the abnormal sound.
Audio event detection, Unsupervised learning.
B.S., Computer Science and Technology, Xidian University, Shaanxi, China
My work mainly focus on spoken keyword spotting. At present, my research is still at the initial stage. So I need to spend more time on literature research and then create my own ideas and implement them. I believe that through my own efforts, I can make great progress.
Automatic speech recognition, deep neural network.
B.S.,School of Architecture and Construction,ChongQing jiaotong Unniversity,Chongqing,China
My research interest is knowledge graph of the natural language processing. Interpretability and reasoning ability are the bottleneck of deep learning. Starting from cognitive intelligence, knowledge graph can bring more cognitive ability to machines.I will continue to learn knowledge in this field and combine it with relevant directions in the laboratory to achieve mutual promotion.
B.E.,School of Information Engineering,HuangShan University,Anhui,China
My research direction is speaker recognition and fault tolerance training. The main research content is to train the optimal speaker recognition model when the data set contains wrong speaker tags.At present, I am still in the exploratory stage, and I hope to achieve success under the guidance of the teacher in the future.
Speaker Recognition,Machine Learning,Deep learning.