Audio gain control method, system and device and storage medium

Abstract

The invention discloses an audio gain control method, which comprises the following steps of: performing activity detection on each frame in an audio to determine the type of a current frame, the typeof the current frame comprising a mute frame and an activation frame; when the type of the current frame is an activation frame, pressing the activation frame into an array; after the activation frame is pressed into an array, determining the amplitude of each audio point in the activation frame, each frame comprising a plurality of audio points; and performing gain control on the audio point according to the amplitude of the audio point and a preset compressor parameter. The invention further discloses an audio gain control system, audio gain control device and a computer readable storage medium. By adopting the embodiment of the invention, the gain of the audio can be automatically controlled, and the sensitivity of audio gain control is improved.

Description

technical field [0001] The present invention relates to the technical field of voice processing, in particular to an audio gain control method, system, device and storage medium. Background technique [0002] In the application scenario of the security camera, or in the voice call, the following problems often occur: the collected volume fluctuates, and it is not stable, causing a very bad experience for the user. To this end, we need some processing means, when the audio sound exceeds the preset loudness threshold, the loudness of the excessive audio frame will be lowered as a whole. But turning down the volume of some frames can easily cause audio distortion. To this end, some methods of controlling the volume are needed, and the academic circles call this type of problem Automatic Gain Control (Automatic Gain Control). In the audio processing industry, the commonly used gain control method first judges the VAD attribute of the current frame by using VAD (Voice Activity ...

AI Technical Summary

Problems solved by technology

[0002] In the application scenario of the security camera, or in the voice call, the following problems often occur: the collected volume fluctuates, and it is not stable, causing a very bad experience for the user

However, this method is relatively rough. In many scenarios (such as security monitoring), the volume value of the signal varies greatly, so the calculated average energy value is difficult to reflect the real situation of the current signal, resulting in insufficient sensitivity of the audio gain control.

Images