Noise reduction method for earphone call and earphone

Abstract

The invention provides an earphone call noise reduction method and an earphone. In one embodiment, the method comprises the steps of receiving a plurality of first speech signals outside a human ear by at least one external microphone; receiving a second voice signal in the human ear through an inner ear microphone; converting the plurality of first voice signals and the second voice signal into afrequency domain; performing beam forming on the plurality of first voice signals to obtain a third voice signal; performing echo cancellation processing and spectrum continuation processing on the second voice signal to obtain a fourth voice signal; and calculating the signal-to-noise ratio of each frequency point in the third voice signal, performing weighting processing on the third voice signal and the fourth voice signal through the signal-to-noise ratio to obtain a fifth voice signal, and converting the fifth voice signal into a time domain and outputting the fifth voice signal. Under the complex acoustic scenes of low signal-to-noise ratio, more speaker interference and the like, the voice of a wearer can be effectively picked up, and other noises can be suppressed.

Description

technical field [0001] The invention relates to the technical field of voice noise reduction, in particular to a noise reduction method for earphone calls and an earphone. Background technique [0002] With the development of the Internet and artificial intelligence technology, earphones have been endowed with multiple functions, including listening to music, voice calls and voice assistants. In practical applications, the earphone microphone will also pick up various noises in the received signal, including relatively stable environmental noise such as subway noise and road noise, as well as non-stationary noise generated by other nearby speakers or broadcasts from shopping malls and high-speed rail stations. These noises will not only make the wearer feel uncomfortable, but also affect the call quality and human-machine voice interaction performance. Therefore, more and more headsets have added functions such as call noise reduction and active noise reduction. [0003] C...

AI Technical Summary

Problems solved by technology

[0005] At present, the commonly used earphone call noise reduction methods are mainly divided into three categories. The first category is to use the voice enhancement algorithm based on the microphone array to enhance the voice of the target direction, and to enhance the voice of the wearer through the adaptive beam technology, but this method is low. Severe performance degradation in signal-to-noise ratios

The second type is noise reduction based on auxiliary sensors, such as vibration sensors or acceleration sensors (such as bone conduction sensors) to extract the wearer's voice. This type of method can effectively extract the wearer's low-frequency voice in the case of low signal-to-noise ratio. , but this type of sensor requires the wearer to wear the headset correctly and the cost is high

The third category is to use machine learning methods to enhance the voice of received signals. This type of method can effectively remove dynamic noise but consumes more computing resources and memory.

[0010] At the same time, the inner ear microphone also has certain limitations:

[0011] 1. There are many signal components including earphone speakers, which require additional echo cancellation processing;

[0012] 2. The medium and high frequency components of the target voice in the signal received by the inner ear microphone are seriously lost, and the sound quality of the medium and high frequency components is significantly worse than that of the signal received by the microphone outside the shell under the condition of high signal-to-noise ratio

Images