Audio processing method and device, intelligent equipment and storage medium

Abstract

The embodiment of the invention discloses an audio processing method and device, intelligent equipment and a computer readable storage medium. The method comprises the following steps: acquiring a to-be-processed audio signal, and dividing the to-be-processed audio signal into a to-be-processed first frequency belt signal and a to-be-processed second frequency belt signal according to an audio frequency belt division value; performing feature extraction on the to-be-processed first frequency belt signal to obtain an N-dimensional low-frequency feature parameter, and obtaining an M-dimensional coefficient feature parameter according to the N-dimensional low-frequency feature parameter and the to-be-processed audio signal; dividing a to-be-processed second frequency belt signal into K frequency belts, and performing feature extraction on each frequency belt to obtain K-dimensional high-frequency feature parameters; and calling a noise optimization model to process the N-dimensional low-frequency characteristic parameter, the M-dimensional coefficient characteristic parameter and the K-dimensional high-frequency characteristic parameter to obtain a P-dimensional gain parameter. The gain calculation processing is carried out on the to-be-processed audio signal through the gain parameter, and the adverse effect caused by the noise audio signal in the to-be-processed audio signal is improved.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to an audio processing method, device, intelligent device and computer-readable storage medium. Background technique [0002] With the continuous development of communication technology, users can perform voice communication anytime and anywhere. Since the communication location is not fixed, there may be noise in the surrounding environment during communication, so that some unnecessary signals are carried in the transmitted audio signal. [0003] In one scenario, the unnecessary signal mainly refers to a noise signal, and the noise signal includes an echo signal and sound signals emitted by other objects (such as a car horn). These noise signals will adversely affect the voice signal in the communication process. For example, the user at the voice signal receiver cannot clearly hear the voice content of the user at the voice signal sender, which reduces the quality of v...

AI Technical Summary

Problems solved by technology

These noise signals will adversely affect the voice signal in the communication process. For example, the user at the receiver of the voice signal cannot hear the voice content of the user at the sender of the voice signal, which reduces the quality of voice interaction.

Images