Microphone array signal processing system and method

Abstract

The invention discloses a microphone array signal processing system and method. The system comprises a FPGA, a CPU, a main I2S, a sub I2S and a microphone array, wherein the FPGA is connected with the microphone array through the sub I2S; the CPU is connected with the FPGA through the main I2S, and the sampling frequency of the main I2S is the integral multiple of the sub I2S. The parallel processing on the multiple microphone array data is realized through the FPGA, the convergence capacity on the microphone data by the system is improved; the audio data is coded through the FPGA , the CPU directly processes the coded data, and the processing speed of the CPU is improved; the main I2S sampling frequency is the integral multiple of the sub I2S, so that the multiple audio data converged by the FPGA can be effectively transmitted to the CPU, and the audio data receiving speed is improved. By using the FPGA, the scheme that the multi-core processor XMOS or the I2Splatform is used for performing the data convergence of the multiple microphone array data in the prior art is overcome, and the system cost and consumption are reduced.

Description

technical field [0001] The invention relates to the field of electrical signal processing, in particular to a microphone array signal processing system and method. Background technique [0002] In recent years, with the maturity of array signal processing technology, microphone arrays are gradually being used in voice signal processing systems such as video conferencing, large stages, vehicle systems, security control, and smart homes, and the application fields are becoming more and more extensive. [0003] In the application process of the microphone array, there is a problem of inputting multiple channels of audio signals, and it is necessary to process the input multiple channels of audio signals. In the process of processing multi-channel audio signals, there are problems such as system complexity, high circuit cost, and inflexible application. [0004] In the prior art, XMOS is often used for multi-channel pulse density modulation (Pulse Density Modulation, PDM) micro...

AI Technical Summary

Problems solved by technology

In the process of processing multi-channel audio signals, there are problems such as system complexity, high circuit cost, and inflexible applications.

[0004] In the prior art, XMOS is often used for multi-channel pulse density modulation (Pulse Density Modulation, PDM) microphone data aggregation, but XMOS is a multi-core microcontroller unit (Microcontroller Unit, MCU) system, when doing multi-channel PDM interface microphone data aggregation, A decimation filter is required to restore the microphone data. This solution has high cost, high power consumption, and complex system. Using a multi-core MCU to complete this work is an overkill.

There are also processor platforms with multi-channel integrated circuit built-in audio bus (Inter-IC Sound, I2S) input to complete the acquisition of multi-channel microphone data, but there are not many processor platforms with multiple I2S inputs at present, such as: Rockchip’s RK3229 and Amlogic’s S905D, etc. The problem with this solution is that the solution is only applied to the corresponding fixed platform. Once it is replaced with another platform, the whole solution cannot be used.

Images