Howling suppression method and system, storage medium and earphone equipment

Abstract

The invention discloses a howling suppression method and system, a storage medium and earphone equipment. The howling suppression method comprises the following steps: extracting feature information in a sound signal picked up by a pickup microphone; inputting the feature information into a preset howling suppression neural network model so as to output path information of transmitting a sound signal of a loudspeaker to the pickup microphone through a howling suppression neural network model; and calculating a target sound signal picked up by the pickup microphone in the sound signals of the loudspeaker according to the path information, and eliminating the target sound signal from the sound signals picked up by the pickup microphone. The method and system do not affected by linear signals or non-linear signals, and can block the generation condition of the howling and suppress the howling.

Description

technical field [0001] The present application relates to the field of sound control, in particular to a howling suppression method, system, storage medium and earphone equipment. Background technique [0002] Encouraged by the market, the auxiliary hearing function of TWS (True Wireless Stereo, True Wireless Stereo) earphones is becoming more and more important. The auxiliary hearing function is to amplify the missing frequency bands of hearing-impaired patients. The distance between the loudspeaker and the pickup microphone in the TWS headset is relatively close, so the sound signal S picked up by the pickup microphone MIC In addition to including the pure acoustic signal S I , also includes a part of the loudspeaker amplified sound signal S L1 , the sound signal S picked up by the pickup microphone MIC After being amplified by the internal algorithm of the earphone, it is sent to the speaker for output. After multiple cycles, the sound signal S L1 Infinite superpositi...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, an adaptive filter is generally used to output the signal S according to the loudspeaker L0 Estimate the sound signal S picked up by the speaker by the pickup microphone L1 , but the adaptive filter can only estimate the linearly changing signal, while the TWS earphone will amplify the relevant frequency band when it realizes its auxiliary hearing function. L0 Non-linear distortion occurs, correspondingly, the speaker output signal S L0 The feedback path to the pickup microphone after being reflected by the components in the TWS earphone housing cannot be accurately estimated, resulting in the adaptive filter being unable to accurately estimate the sound signal S picked up by the speaker by the pickup microphone L1 , it is impossible to eliminate the sound signal S picked up by the pickup microphone MIC The sound signal S in L1 , which makes the existing howling suppression scheme less effective and brings bad experience to users

Images