Audio data processing method, device, terminal and system

Abstract

The embodiment of the invention discloses an audio data processing method, apparatus, terminal and system. The audio data processing method includes the steps: acquiring the audio data after space filtering; performing first Wiener filtering and second Wiener filtering on the audio data to obtain the first filtering data and the second filtering data, wherein the noise suppression degree of the first Wiener filtering is greater than the noise suppression degree of the second Wiener filtering; and utilizing the first filtering data to perform stop determination on the second filtering data, andaccording to the stop determination result, performing data processing on the second filtering data. The audio data processing method can respectively perform different degree of Wiener filtering according to different demand for voice activity detection and automatic voice recognition, can guarantee the accuracy of automatic voice recognition, can avoid the influence of interference on voice activity detection, can more accurately detect the voice activity state, can reduce the feedback time delay of voice interaction, can improve the response speed for the voice command, and can bring better use experience for the user.

Description

technical field [0001] The present invention relates to the technical field of data processing, in particular to an audio data processing method, device, terminal and system. Background technique [0002] Intelligent voice interaction is an important branch of the field of artificial intelligence. Realizing free and accurate intelligent voice interaction will greatly liberate people's hands and obtain more free information flow and control with the physical world. Intelligent voice interaction mainly includes near-field voice interaction and far-field voice interaction. Near-field voice has been greatly developed in the past two decades. The current near-field voice recognition rate is close to the human voice recognition rate, but the more free interaction should be far-field voice interaction. The so-called far-field voice interaction means that there is a certain distance between the speaker and the interactive device, which expands the free space of the speaker, but thi...

Images