Abstract
In this paper, a speech recognition system is designed and the relevant hard-ware environment is built through Zynq FPGA AX7020 platform. In this system, the feature extraction of speech signals is based on MFCC (mel-frequency cepstrum coefficients). The algorithm in this system includes Pre-emphasis, Framing and Windowing, FFT, MFCC Parameter Calculation, VED Endpoint Detection and DTW Operations processes commonly used in speech system design. In this paper, the speech recognition algorithm is implemented on Zynq FPGA platform and also simulated in Matlab. At the same time, audio module AN831 is adopted in the hardware part to realize the sound acquisition function. Through the comparison of experimental data, it is found that the average recognition rate in Matlab simulation is 86.67%, while in Zynq FPGA platform test, the recognition rate is 80.67%. Using Zynq FPGA AX7020 as a platform to realize speech recognition, the recognition rate is higher than the traditional speech recognition technology.
Export citation and abstract BibTeX RIS
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.