|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | Pinyin N-best search in mandarin speech recognition 显示文摘 | WANG Zuoying LU Dajin | 1999 | Acta Electronica Sinica1999，27，4： | 1 |
| 2 | A new multi-model coding algorithm based on MBE and spectral amplitude correlation between successive frames显示文摘The researches on spectral amplitude correlation between successive frames based on MBE (Multi-Band Excitation) are presented and a new MBE coding algorithm of muitimodels is proposed. It has been shown that the magnitude spectrum in MBE coding algorithm can be estimated in very high accuracy by using a simple linear predictive model of the spectral amplitude correlation between successive frames. This model employs oniy one gain coefficient and one predictive coefficient. The accuracy of estimated magnitude spectrum in this model is higher than that of the conventional MBE algorithm using the 10-th order LPC (Linear Predictive Coding) Inodel without the spectral amplitude correlation between successlve frames.This model is superior to the conventional MBE coding model in volced speech, but it is iess effective in unvoiced speech, so that the conventional MBE coding algorithm is still used for unvoiced speech. The new multi-model coding algorithm improves the quality of coding speech,and obtains good | PAN Shengxi LIU Jia WANG Zuoying LU Dajin(Dept. of Electronic Engineering, Tsinghua University Beijing 100084) | 1998 | Chinese Journal of Acoustics1998，17，3： | 1 |
| 3 | Adaptive sensor management in multisensor data fusion system 显示文摘 | He you Yang Zhi Mao Shiyi Lu dajin | 1999 | Chinese Journal of Electronics1999，8，2： | 1 |
| 4 | Two thresholds track-track association algorithm in distributed multi-sensor data fusion system 显示文摘 | He You Lu Dajin Peng Yingning | 1997 | Journal of Electronics1997，19，6： | 1 |
| 5 | Duration and its application in continuous speech recognition显示文摘DDBHMM (Duration Distribution Based Hidden Markov Model) solved the defects of traditional HMM (Hidden Markov Model). Based on DDBHMM, the problem of how to effectively utilize the duration information is studied in detail. The approach on estimating the duration distribution is introduced first, then the data file is classified according to the speak rate. The recognition experiment shows that, the duration information behaves best on the data of low speak rate, behaves normal on the data of medium speak rate and has little effect on the data of fast speak rate. Therefore, the most importance of duration is that by it the more accurate state segmentation point could be obtained and then the recognition rate can be improved. At the same time, the robustness of the system to speaking rate is improved with the employment of the duration information. Furthermore, the method of classified duration and normalized duration is also put forward and studied in detail, it shows that both of the two method can improve the performance. In order to study the dependency between the duration, the method of using the Bigram of the duration is proposed and analyzed. At last, the approach of post processing duration is studied, it shows that only based on DDBHMM, and utilizing the duration information synchronously in the recognition process, then the performance can be improved greatly. | ZHAO Qingwei XIAO Xi WANG Zuoying LU Dajin (Department of Electronic Engineering, Tsinghua University Beijing 100084) | 2000 | Chinese Journal of Acoustics2000，19，3： | 0 |
| 6 | Decidualization-derived cAMP regulates phenotypic and functional conversion of decidual NK cells from CD56dimCD16− NK cells显示文摘In early pregnancy,the decidua is populated by a large number of natural killer cells(NKs),most of which are CD56bright with a strong capacity to secrete cytokines,differing from peripheral and cord blood NK cells(pNKs and cNKs).1,2,3 It is generally believed that uterine NKs are recruited from the peripheral blood and are further educated to develop a decidual NK cell(dNK)-like phenotype in the microenvironment at the maternal-fetal interface.4 However,the defined mechanisms involved in this education remain unclear. | Xueling Jin Liyuan Cui Weijie Zhao Xinyi Li Lu Liu Yanhong Li Qiang Fu Dajin Li Min Yu Meirong Du | 2021 | Cellular & Molecular Immunology2021，18，6： | 0 |
您所在的机构暂未开通此部分的功能，请联系管理员