In order to improve the voice quality of network English teaching system, the paper operated the study on the improvement of the characteristic parameter of MFCC and LSP, which reduced noise, optimized the voice quality and improved the accuracy of voice judgment in some extent. How to improve the quality of voice identification is the key to optimize the voice quality of network English teaching system. The study first analyzed the key factors of improving voice quality from the following three aspects, which are the preprocessing of voice signal, the extraction of parameters of voice characteristics and the measurement of similarity. And then took the parameters of voice characteristics as the entry point and finished the parameter extraction of voice characteristics by combining MFCC and LSP. The experiment shows that such method not only restores the voice reality of speakers effectively, but also reduces the misjudgment rate of voice matching. The above functions of such method make it own some research value.