| 1 | Structural organization and scale-free properties in Chinese Phrase Networks显示文摘Monkey language models are defined for Chi-nese Phrase Networks, and scale-free features of Chinese Phrase Networks are uncovered. It is pointed out that the ratio of average degree to the total number of nodes ( k /N ) is close to a constant. Simulation for the evolution of phrase networks indicates that one of the important reasons for power law distributions is the word selection frequency, which, when tuned aptly, can make the monkey language present similar statistic traits as that of natural languages. Power law tails emerge at large k, and the exponent is about 6. Comparison between monkey model and natural language shows that humans are able to use Chinese words resources in more effective and compact ways to express their inten-tions. All the results demonstrate an important fact that the least effort principle is the basis of Chinese Phrase Networks. | LIYong WEILuoxia NIUYi YINJunxun | 2005 | Chinese Science Bulletin2005，50，13： | 6 |
| 2 | A Perceptual Audio Representation for Low Rate Coding Based on Sines+Noise Modeling显示文摘This work is concerned with the develop-ment and optimization of an efficient (which allows high compression ratios) and flexible (which allows scalability)signal model for perceptual audio coding at low bitrates.A novel, complementary two-part model for audio consist-ing of sines+ noise (SN) is presented. The SN model uses a sinusoidal model that explicitly takes into account the human hearing system by using psychoacoustically based matching pursuits. This technique iteratively extracts si-nusoidal components according to their perceptually im-portant signal-to-mask ratio (SMR). The second modeling stage is for noise-like components. The SN model uses the equivalent rectangular bandwidth (ERB) noise model;that is based on observations that for noise-like signals,energy in the ERBs describes the underlying signal with perceptual accuracy. The SN model has an intuitive inter-pretation in terms of discrete fourier transform (DFT) and can be efficiently implemented via the fast fourier trans-form (FFT). Informal listening tests demonstrate that the synthesized (sines + noise) signal is almost perceptually identical to the original. A compression ratio of typically 16 to 19.5 can be readily reached with SN model. | AL-MoussawyRaed YINJunxun HUANGJiancheng | 2003 | Chinese Journal of Electronics2003，12，3： | 0 |
您所在的机构暂未开通此部分的功能，请联系管理员