Research Of Speaker-Recognition System Based On Embedded Linux

In recent years, the speaker recognition technology gradually matured, as a biometric technology, speaker recognition has its unique advantages, such as the voice of non-contact and naturally, users find it easier to accepted. As the voice of the media's ,it has own special advantages, the adoption of voice authentication technology is rapidly moving toward practical use, highlighted the huge market potential, speaker recognition technology has become a new competitive high-tech industry.With the computer software and hardware technology semiconductor technology,electronic communications technology and network technology, as well as the continuous development of embedded technology and updated its powerful performance and portability greatly enhanced. A large number of real-time data collection, filtering, processing in low-power, small size of the embedded device to complete. ARM processors, today's orders because of their special structure and high efficiency to enable it to quickly compile the implementation of Voice Recognition Algorithm to meet today's digital signal processing and high real-time requirements. High-performance ARM processor voiceprint identification system at the same time as a result of voiceprint recognition technology with the convenience, economy, accuracy and portability of embedded systems, mobility and other advantages by the worldwide attention, has been widely applied to the people of the daily life, and become the main trend of scientific and technological development. This paper studied speech recognition system based on ARM technology. It introduces the process of speech recognition realizing with ARM chip in theory analysis, system collectivity design, system hardware design and system software design. This paper also introduces VQ model and speech digital signal processing associated with speech recognition .Some algorithms are introduced concretely in this paper such as Mel reverse spectrum parameters ,LMS adapting algorithm, LBG algorithm etc. This paper analyses recognition algorithms and do some improvement to build a high performance and low operation recognition system of particular person words. All of these are experimented and based on this theory. The hardware part of this system mainly include memory system, Audio Processor Chip. The software part of this system mainly include boot loader, kernel, root file system etc. Finally, this paper talks about speech recognition technology based on ARM by a real speech recognition system.According to the experiment, this system can recognize speech . For further development of using ARM Embedded Processor Portable Speaker Recognition equipment to lay the groundwork...