Researching And Implementing Of Speech Recognition System Based On The Android Platform

With the rapid development of mobile Internet and intelligent human computer interaction technology, convenient and efficient human-computer interaction has been widely concerned. As the most natural human-computer interaction, speech recognition technology can effectively improve the low speed, easy error and other defects of the traditional manual keyboard, So that the application of speech recognition technology in human computer interaction is more and more large, and gradually become a competitive value of the emerging high-tech industry hot spots.So far, Speech recognition technology in the mobile intelligent terminal application develops rapidly, and people can control the terminal directly through natural speech, such as dialing the phone, searching information, opening applications, etc. At the same time, people also began to try applying the speech recognition system based on mobile intelligent terminal to some control areas, in order to improve the efficiency of human-computer interaction and achieved good results. However, the technology is still in the initial stage, so that it has extensive application value and development prospects for the recognition technology applied to different fields. Based on this, this paper will use CMU Sphinx toolbox to research the realization of speech recognition technology on the Android platform, and try to apply it to the Electronic Shelf Label management system, in order to improve the operation of the portability and efficiency.Firstly, the paper will introduce and analyze the key technology of speech recognition system in detail, and put forward a kind of combining wavelet transform and based on short-time energy zero difference threshold detection method to improve noise ratio of the speech and endpoint detection accuracy when the recognition environment is very noise. Then, the paper will focus on designing and implementing the speech recognition system based on the Android platform, and applying it to the ESL management system. Finally, The system will be tested to verify its effectiveness and performance.Through testing identification of the rate, anti noise and efficiency of operation, the results show that the ESL voice management system can manage the ESL in a high recognition performance and human computer interaction efficiency. The success of the ESL voice management system not only can conducive the promotion and application of ESL, but also can provide a reference case and ideas for the speech recognition technology based on the Android platform applying to other control areas.