New Speech Noise Reduction Recognition System Based on Spatial Filtering Technology and CI1103 Speech Module

Abstract

With the continuous development of science and technology in recent years, there are more and more ways of human-computer interaction, and its technology is becoming more and more mature. Among them, the human-computer voice interaction mode occupies an important position, which is gradually integrated into our lives. However, through reviewing and summarizing previous studies, it is found that speech recognition still has some shortcomings in noise reduction processing [2]. Therefore, to design a system that is immune to ambient noise and can perform voice human-computer interaction more accurately, based on spatial filtering technology and CI1103 speech recognition module, a new type of noise reduction speech recognition system using a unique anti-noise processing microphone is proposed. In this new system, combined with the built-in anti-noise filter of CI1103, the noise reduction process during speech recognition is realized with the help of the spatial filtering technology where the new digital signal is used to process IC “BU8332KV-M” [3]. Its peripheral control interfaces such as built-in CPU core and Audio Codec module with high performance and low power, integrated multiple UART, IIC, SPI, PWM, and GPIO can develop all kinds of cost-effective and single-chip intelligent voice products. Experimental results show that the new noise reduction speech recognition system has higher speech recognition accuracy than that of the traditional speech recognition systems in noisy environments, which has the function of rejecting false recognition, and is suitable for various speech systems with environmental noise interference.