Special Session on Speech and Language Processing for Han Dialects

Chair: Dr. Ming-Shing Yu, National Chung-Hsing University, Taichung

  Email: msyu@nchu.edu.tw

In the past decades, there is a rich tradition of research focusing on Chinese speech and language processing, mainly on Mandarin (and Pu-Tong-Hua). Yet there are many dialects of Chinese. These languages also carry wisdoms accumulated by the Chinese people. Some important dialects, for example, are Guangdong Hua in Hong Kong, Taiwanese and Hakka in Taiwan, etc. The purpose of this special session is to present the recent advances in the areas of speech and language processing for the branches of Chinese other than Mandarin and Pu-Tong-Hua. We invite papers on, but not limited to, the following topics on Han Dialects:

Speech and Natural Language Processing
Research on Writing Systems and Related Computer Processing
Research on TTS Systems
Research on Computer Assisted Instruction or Learning
Research on Speech Recognition
Research on Syntax and Semantic Analysis
Research on Input Systems
Research on Corpus Development
Other Applications including Speech and Natural Language Processing


Special Session on New Paradigms in ASR

Prof. Hsiao-Chuan Wang, National Tsing Hua University, Hsinchu

Prof. Yuan-Fu Liao, National Taipei University of Technology, Taipei


Email: yfliao@ntut.edu.tw

Although we have learned a great deal about how to build data driven-based automatic speech recognition (ASR) systems for almost any spoken language, the performance of existing state-of-the-art technology is somewhat fragile in comparing with human speech recognition (HSR). For example, the state-of-the-art ASR systems usually give much higher error rates even in clean environments and the accuracy often declines dramatically in adverse conditions.

To bridging the gap between human and machine a paradigms shift in ASR may be necessary. Recently, some new ASR approaches have been proposed and their experimental results are shown to be promising. Good examples are detection-based speech recognition, information retrieval methods for automatic speech recognition, and so on.

The purpose of this special session is to investigate and present some new ideas beyond the state-of-the-art ASR approaches. We invite papers on, but not limited to, the following topics:

Research on new paradigms in ASR
Research on new speech features
Research on new acoustic, prosodic and language modeling
Research on new algorithms for ASR
Other related topics in new paradigms in ASR


Paper submission deadline: September 2, 2010

(Please submit your papers to the special session chairs directly.)

TEL:+886 6 2096455
FAX:+8866 2381422