High-frequency optimization method and device for audios and medium

Abstract

The invention discloses a high-frequency optimization method for audios. The method comprises the following steps of: carrying out high-frequency reconstruction on an obtained low-sampling-rate audioto obtain a high-sampling-rate audio; calculating to obtain a low-sampling-rate frequency domain signal and a high-sampling-rate frequency domain signal; under the condition that it is judged that thelow-frequency difference between the low-sampling-rate frequency domain signal and the high-sampling-rate frequency domain signal meets a preset condition, performing replacement processing on the low frequency of the high-sampling-rate frequency domain signal according to the low-sampling-rate frequency domain signal; carrying out first smoothing processing on the high frequency of the high-sampling-rate frequency domain signal; performing fundamental frequency search on the low-sampling-rate frequency domain signal; and under the condition that a fundamental frequency exists, performing secondary smoothing processing on the high frequency of the high-sampling-rate frequency domain signal after the primary smoothing processing according to the fundamental frequency. The invention furtherdiscloses a corresponding optimization device and a medium. High-sampling-rate audios obtained after high-frequency reconstruction of low-sampling-rate audios are further optimized, so that the overall generation quality of the audios is greatly enhanced.

Description

technical field [0001] The present invention relates to the technical field of communication, in particular to an audio frequency optimization method, device and medium. Background technique [0002] In the field of audio and speech technology, high-sampling rate audio has better detail information, while low-sampling rate audio lacks detail information, and the sense of hearing will be worse than high-sampling rate audio. In many cases, due to insufficient hardware conditions, or high-sampling-rate audio requires more traffic for storage and transmission, only low-sampling-rate audio can be obtained. Therefore, it is a key technology to perform high-frequency reconstruction on low-sampling-rate audio. [0003] Existing high-frequency reconstruction methods include high-frequency reconstruction methods based on traditional machine learning, high-frequency reconstruction methods based on deep learning, and so on. However, in the process of implementing the present invention...

AI Technical Summary

Problems solved by technology

However, in the process of implementing the present invention, the inventors found that the prior art has at least the following problems: the effect of the high-sampling rate audio generated by the high-frequency reconstruction method based on traditional machine learning and the high-frequency reconstruction method based on deep learning is unstable, extremely Susceptible to noise interference, and the high frequency part often has obvious intermittent phenomenon between frames

Images