2007年清华大学信息技术研究院系列学术报告9

Biography

Lin-shan Lee received a B.S. from Taiwan University in 1974, and a Ph.D. from Stanford University in 1977. He has been a professor of Taiwan University since 1982. He also holds a joint appointment with Academia Sinica at Taipei as a research fellow since 1985. His research areas include digital communications and spoken language processing. He developed quite several earliest versions of Chinese spoken language systems in the world, including text-to-speech systems, natural language analyzers, voice dictation systems, spoken document retrieval systems, and spoken dialogue systems. His recent research work in this area is focused on spoken language processing technologies for network information access, including user interface, content analysis and interactions between user and the content. He was a member of the Permanent Council of International Conference on Spoken Language Processing (ICSLP) 1994-2004, and elected as a Board member of International Speech Communication Association (ISCA) 2001-2009. He will be the general chair of IEEE ICASSP 2009 at Taipei. He also served as the Vice President for International Affairs (1996-1997) and the Awards Committee chair (1998-1999) of IEEE Communications Society. He was elected IEEE Fellow in 1992.

Abstract

The network environment introduces new challenges and new opportunities to spoken language processing technologies. The many new network-based applications, the huge resources offered by the networks, the large number of network users and the complicated user conditions are some examples. The spoken language processing technologies under network environment may include at least those for user interface, for network content analysis, and for the interaction between the user and the content.

This talk very briefly presents a few example works recently done at Taiwan University towards this direction, including robust speech recognition, spontaneous speech processing, Prosodic modeling, spoken document retrieval and understanding, spoken and multi-modal dialogues, and distributed speech recognition. Basic knowledge about spoken language processing is assumed for the audience.