Human Computer Speech Interaction Research Group (hcsi): Department of Computer Science and Technology, Tsinghua University


Lab Name and Affiliation

Human Computer Speech Interaction Research Group (hcsi): Department of Computer Science and Technology, Tsinghua University

Lianhong Cai, Jia Jia, Mingxing Xu

Lab Director (or Principal Investigator)

Lianhong Cai

Prof. Cai Lianhong graduated from Department of Automation, Tsinghua University at 1970. Now she works in Department of Computer Science and Technology, Tsinghua University, and is a professor from 1994 and a doctor supervisor from 1998. Currently, she is the leader of the Institute of Human Computer Interactive and Media Integration, the member of Multimedia Committee of Chinese Graphics and Image Society, the member of Chinese Acoustics Society, and the member of IEEE.

Her major research interests include Human Computer Speech Interactive, Speech synthesis and TTS, Speech Corpus and multimedia technology. She has taught "The Digital Processing of Speech Signals", "Multimedia Computer Technologies" and so on. She has been undertaking several the national key-tech tasks, 863 high technology projects and National Natural Science Foundation projects. She has been awarded Scientific Progress Prizes and the Invention Prizes from the Ministry of Mechanism and Electronics and the State Education Commission, and some patents. She has published tens of papers and several scientific literatures

Lab Introduction

HCSI Group is an important part of the Speech Laboratory in the computer science and technology department of tsinghua University. For a long time, the group has made a great effort on the study and the development of Acoustical model, Prosody learning and simulation, Automatic text analysis, Prosody description language for Chinese Speech, Digital coding/decoding technology, Multimedia communication technology and other associated kernel technologies. In 1989, the group firstly developed a database reading system with speech synthesis technology in China. It makes a great progress in the application of the Chinese speech synthesis system.
"Make computer speak naturally and smoothly just like a human-being" is always the goal of the group. The group has made a great effort on the research of the new models and technologies, and applied them to the TTS system. During recent years, the group has developed several practical system, such as the time domain TTS system (1992) and the TH-SPEECH system (1993), etc. At present, the TTS system has formed several versions. It can be run in different operating systems, DOS, Window95/98/NT, and Unix, and has been widely used in querying system of telephone, soundable electronic dictionary, manuscript verifying system, newspaper reading system, etc. Now, all members of the group are working hard to make more progress of TTS system, and make it intelligentized and more useful in network field.
The group has established the abroad affiliations with many universities and research institutions all over the world, and delivered many high-quality papers on many national and international journals and conferences. The members in group has been invited to visit many countries and regions, such as USA, Germany, Singapore, Hongkong, Taiwan, etc. Nowadays, the group still keeps the friendly cooperations with Lucent company, Dialogic company, Siemens company, Intel company and Microsoft company.

Lab Contact E-mail