Abstract: The present disclosure provides a magnetic random-access memory, comprising: an antiferromagnetic layer; a magnetic tunnel junction disposed on the antiferromagnetic layer and comprising a ferromagnetic layer disposed corresponding to the antiferromagnetic layer; wherein the ferromagnetic layer of the magnetic tunnel junction has in-plane magnetic anisotropy, and an exchange bias field is formed between the antiferromagnetic layer and the ferromagnetic layer by an annealing process. A direction of the exchange bias field is changed by a spin orbit torque, thereby changing a direction of a magnetic moment of the ferromagnetic layer and realizing data writing. The present disclosure can improve a thermal stability of the i-MTJ and reduce a lateral dimension of the i-MTJ, thereby improving a storage density of the magnetic memory.

Abstract: A method, an apparatus and a system for testing a bone conduction communication device is disclosed. A vibration exciter is disposed at a wearing position of an artificial head model where the bone conduction communication device is located. The method comprises: transmitting a test audio to first and a second digital signal processors; transmitting an audio signal processed by the first digital signal processor to the vibration exciter, and simulating vibration of a bone at a corresponding part by which human generates sound using the vibration exciter, to generate a vibration signal; transmitting an audio signal processed by the second digital signal processor to an artificial mouth and simulating human sound production to generate an acoustic signal; collecting the vibration signal and the sound signal using the bone conduction communication device, and testing the bone conduction communication device according to the collected vibration signal and sound signal.

Abstract: An audio signal processing method is disclosed. The method comprises the following steps: acquiring a first audio signal received by a bone conduction sensor and a second audio signal received by the microphone; determining a frequency division value according to the first audio signal and the second audio signal; and processing the first audio signal and the second audio signal according to the frequency division value to obtain an audio output signal of the electronic device. An electronic device and a computer readable storage medium are also disclosed, which achieves the effect of improving the quality of audio signals collected by the electronic device.

Abstract: An audio signal processing method including the following steps: acquiring a bone conduction audio signal collected by a bone conduction vibration pickup device; acquiring a low frequency transfer function which matches the bone conduction audio signal and a high frequency transfer function corresponding to the low frequency transfer function, wherein the low frequency transfer function and the high frequency transfer function are corresponding transfer functions between the bone conduction audio signal and an air conduction audio signal which correspond to the same voice; and performing frequency domain extension on the bone conduction audio signal on the basis of the low frequency transfer function and the high frequency transfer function, and taking the extended initial audio as output audio. The effect of improving the integrity pf a voice signal collected by a terminal device is achieved.

Abstract: A signal processing method and device of a MEMS microphone and a MEMS microphone are disclosed. The method comprises: acquiring a first electrical signal obtained by converting a received optical signal by an optical sensor disposed near a sound hole of the MEMS microphone; acquiring a second electrical signal output by the MEMS microphone; and judging that the second electrical signal is an interference signal when within a substantially overlapping time range, the first electrical signal acquired reaches a preset first threshold value and the second electrical signal acquired meets a preset condition.