全部

1·The imitating system by MATLAB experiments on the text composed of vowel, voiceless consonant and voiced consonant. The result shows that the recognition rate is over 98.5%.

用MATLAB对说话人识别系统进行仿真实验，选择由元音、轻辅音和浊辅音组成的本文，经过测试，识别率达98.5 %。

2·According to a comprehensive investigation, 17 common characteristics like the convergence of voiced consonant and voiceless consonant are found in pronunciation in Gan dialect.

通过全面考察得知：江西赣方言语音有“全浊声母与次清声母合流”等17个方面的共同特点；

3·While singing Italian songs, Chinese often cannot differentiate voiceless consonant from voiced one, but read consonant cluster according to Chinese habit.

中国人在演唱意大利语歌曲时，通常不能区分清、浊辅音，按汉语的习惯来读复辅音。

4·A consonant can either be unvoiced or voiced.

辅音可发声也可不发声。