Audio noise reduction method and device, equipment and medium

Abstract

The invention relates to an audio noise reduction method, device and equipment and a medium, which are applied to voice control scenes such as elevators and intelligent automobiles and are used for carrying out noise reduction processing on audio of each object and optimizing a voice recognition process. The method comprises the steps of obtaining voice audio of a target object; acquiring a voice audio of the target object; based on a frequency domain and barker domain transformation relationship, determining a target barker band corresponding to the amplitudes of a target voice signal at the plurality of preset frequencies, the target voice signal being any frame of voice signal of the voice audio; determining an audio feature of the target voice signal by using a matrix formed by the determined target barker bands; inputting the audio features of the target voice signal into a noise reduction network model to obtain a sound-to-noise ratio matrix corresponding to the target voice signal; and determining the denoised target voice signal based on the sound-to-noise ratio matrix and the amplitudes of the target voice signal at the plurality of preset frequencies.

Description

[0001] This application claims the priority of the Chinese patent application submitted to the Intellectual Property Office of the People's Republic of China on July 3, 2020, the application number is 202010635457.5, and the invention title is "A processing method, device, equipment and medium for audio signals". The entire contents are incorporated by reference in this application. technical field [0002] The present invention relates to the field of speech processing, in particular to an audio noise reduction method, device, equipment and medium. Background technique [0003] In the field of voice control, usually one user issues a voice control command at the same time, and the specific command intent is determined according to the collected audio of the user. However, in scenarios where voice control permissions are shared (for example, elevator scenarios), multiple users have voice control permissions. If multiple users issue voice control commands at the same time, th...

AI Technical Summary

Problems solved by technology

Separate each user's voice control command from the collected audio. The quality of the audio is poor, making it impossible to determine the specific command intent of each user. It is also necessary to confirm that multiple users issue voice control commands one by one, which makes the processing efficiency low.

Images