Active noise reduction method, device and system based on deep learning

Abstract

The invention provides an active noise reduction method, device and system based on deep learning. The method comprises: acquiring audio data; based on the audio data, utilizing a trained noise reduction classification model to obtain an audio type result of the audio data; and performing noise reduction processing on the audio data according to the audio type result. Based on deep learning, active noise reduction is carried out on a sound source with a noise elimination requirement, and noise reduction is not performed on a sound source which may contain useful information. The problems thatthe capacity of an active noise reduction algorithm is not high, complete voice information cannot be obtained after noise reduction or a user loses useful information are solved, signals with noise reduction requirements are eliminated, original information of signals to be reserved can be reserved, and then the situation that the user loses useful information due to active noise reduction is avoided.

Description

technical field [0001] The present invention relates to the technical field of sound, and more specifically relates to an active noise reduction method and device. Background technique [0002] Ambient noise is an important factor affecting the sound quality perception of the earphone wearer. Wearing earphones to listen to audio or perform voice communication in a high-noise environment will cause the user's listening quality to decline, and more seriously, the user's hearing will be damaged. Therefore, earphone noise reduction has attracted the attention of earphone designers and users. At present, there are two methods for earphone noise reduction: passive noise reduction and active noise reduction. Passive noise reduction generally achieves physical signal isolation between the earphones and the outside world through the structure and material design of the earphones, such as in-ear designs. Active noise reduction uses analog or digital signal processing methods to offs...

AI Technical Summary

Problems solved by technology

The present invention provides an active noise reduction method, device, system, and computer storage medium based on deep learning, which can purposefully select sound sources, and perform active noise reduction on sound sources that require noise reduction, which is useful for The sound source of the information does not perform noise reduction, and the ability to solve the active noise reduction algorithm is not high. After the noise reduction, the complete voice information cannot be obtained or the user loses useful information. The signal can retain the original information, so as to prevent users from losing useful information due to active noise reduction

Images