Audio processing method and device, computer equipment and storage medium

Abstract

The embodiment of the invention discloses an audio processing method and device, computer equipment and a storage medium. The audio processing method comprises the steps of: obtaining positive samplefrequency spectrum characteristics of positive sample audio, and obtaining negative sample frequency spectrum characteristics of negative sample audio; determining the predicted audio gain of the negative sample audio based on the sample gain prediction model and the negative sample spectrum characteristics; determining a sample audio gain of the negative sample audio according to the positive sample frequency spectrum feature and the negative sample frequency spectrum feature, adjusting a model loss function of a sample gain prediction model according to the sample audio gain or the prediction audio gain, and training the sample gain prediction model according to the adjusted model loss function to obtain a target gain prediction model, wherein the target gain prediction model is used forpredicting the target audio gain of the target audio. By adopting the audio processing method and device, the prediction accuracy of the prediction model can be improved.

Description

technical field [0001] The present application relates to the field of computer technology, and in particular to an audio processing method, device, computer equipment and storage medium. Background technique [0002] With the rapid development of Internet technology and the popularization of handheld recording devices, the audio data that people can obtain has exploded, but the audio quality of the obtained audio data is low, and a large amount of audio data contains noise. In order to suppress the noise in the audio data, the audio data needs to be amplified, and an artificial intelligence method can be used to train a gain prediction model, which is used to predict the gain information of the audio data. [0003] The loss function of the gain prediction model will directly determine the training effect of the model. Currently, the loss function of the model is determined before the model starts training. Once the loss function is determined, the loss function will not cha...

AI Technical Summary

Problems solved by technology

[0003] The loss function of the gain prediction model will directly determine the training effect of the model. Currently, the loss function of the model is determined before the model starts training. Once the loss function is determined, the loss function will not change during the entire model training process. However, participating in model training There may be large differences between a large number of samples, and the fixed loss function cannot match the training samples with large differences, resulting in low prediction accuracy of the trained prediction model

Images