Analysis of the impact of data resolution on the speaker recognition effectiveness in embedded fixed-point systems

This paper presents an analysis of issues related to the resolution of the biometric data of the speech signal processing implemented using fixed-point digital signal processor. Implementation calculations were made using the module with the commissioning of the TMS320C5515 processor, which can be programmed with the use of Code Composer Studio (CCS) environment. We shown features of C5515 digital signal processor that can be used for efficient implementation of algorithms for speech processing. Finally, the influence of data resolution on the accuracy of the calculations and the efficiency of identification is presented.