Signal sending method and device and electronic equipment

Abstract

The embodiment of the invention provides a signal sending method and device and electronic equipment, and the method comprises the steps: obtaining a voice signal which is collected by a microphone array which carries out the voice collection of an interaction site; based on a preset beam forming algorithm, performing beam generation on the acquired voice signal to obtain a plurality of voice beams with different beam directions; obtaining far-end sound source information, wherein the far-end sound source information is used for indicating the direction of a far-end sound source relative to the microphone array in the interaction site; and determining a voice beam with a beam direction different from the direction indicated by the far-end sound source information from the plurality of voice beams as an available beam. Through the scheme, the interaction quality of the voice interaction system can be improved.

Description

technical field [0001] The invention relates to the technical field of speech processing, in particular to a signal sending method, device and electronic equipment. Background technique [0002] like figure 1 Shown is a schematic diagram of a traditional voice interaction system. In a traditional voice interaction system, a single microphone is generally used to collect the voice signal of the local interactive scene. After the single microphone collects the voice signal of the local interactive scene, the collected voice signal can be firstly subjected to echo cancellation, noise reduction and automatic gain. After processing, after encoding, it is transmitted to the interactive terminal through the network. [0003] However, in some occasions, when both sides of the interaction speak at the same time, that is, when there are local audio sources and remote audio sources at the interactive site, the single microphone at the interactive site will simultaneously collect the ...

AI Technical Summary

Problems solved by technology

[0003] However, in some occasions, when the two sides of the interaction speak at the same time, that is, when the local audio source and the remote audio source appear at the interactive site, the single microphone at the interactive site will simultaneously collect the voice signals of the local audio source and the remote audio source.

At this time, after performing echo cancellation and other processing on the voice signal collected by the single microphone, the voice signal from the local audio source will be seriously damaged, and the speaker at the interactive end cannot clearly play the voice signal of the local audio source, resulting in poor interaction quality.

Images