Feature parameters extraction is critical for speaker recognition research. The paper presents the function of pitch, formant and Mel frequency central coefficient (MFCC) in speaker recognition. It can increase the identification rate effectively for feature parameter sorts the speech corpus. Using Euclid Distance to compare feature parameters is very effective.

The anisotropic silicon (Si) etching characteristics of Si (100) in 25 wt.% tetramethyl ammonium hydroxide (TMAH) solutions containing Triton X-100 and isopropyl alcohol (IPA) were studied. The etch rate, convex corners and roughness of the etched surface were investigated. In this reported work two central goals, a mirror-like surface finish and a high reduction of undercutting, have been achieved. The best etched result was obtained in 25 wt.% TMAH + 0.25%v/v Triton + 16%v/v IPA, which has minimum convex corner undercutting and a smooth etched surface (R a = 1 nm). This study is useful for engineering applications where the fabrication of microstructures for high-quality devices should simultaneously contain smooth surfaces on a large area and less convex corner undercutting.

The key generation algorithm of AES was introduced; the weaknesses of the key generation design of AES were investigated. According to the key demand put forward a kind of new design idea, and this designing strategy was developed, which can be used to improve the key generation algorithm of AES. An analysis shows that such improvement can enhance the safety of the original algorithm without reducing its efficiency

No abstract