Method and apparatus for switching main microphone, voice detection method and apparatus for microphone, microphone-loudspeaker integrated device, and readable storage medium

An apparatus for switching a main microphone of a plurality of microphones, comprising an acquiring unit, a preprocessing unit, a first calculating unit, a second calculating unit, and a switching unit;

• the acquiring unit is configured for acquiring first audio data for playback by a loudspeaker, and acquiring second audio data acquired by each microphone and noise frequency domain energy corresponding to each microphone;
• the preprocessing unit is configured for preprocessing the first audio data and the second audio data, obtaining corresponding first frequency domain data and second frequency domain data;
• the first calculating unit is configured for calculating first frequency domain energy of the first frequency domain data and second frequency domain energy of the second frequency domain data;
• the second calculating unit is configured for selecting a candidate main microphone from the plurality of microphones according to the noise frequency domain energy and the second frequency domain energy corresponding to each microphone when the first frequency domain energy is less than a first preset threshold;
• and
• the switching unit is configured for switching the candidate main microphone as the main microphone when the candidate main microphone is different from the current main microphone.

A voice detection method for a microphone, comprising:

• acquiring a local audio signal by N microphone at corresponding angles to obtain corresponding N second audio data, wherein each second audio data comprises an audio frame;
• performing voice activity detection on the N second audio data, determining X second audio data to be voice frames, and determining Y second audio data to be noise frames, wherein X plus Y is equal to N, both X and Y are integers greater than or equal to 0, and N is an integer greater than 0;
• when both X and Y are not equal to 0, detecting whether the N second audio data contain a fundamental frequency of voice and a harmonic;
• and
• when any one of the N second audio data contains the fundamental frequency of voice and the harmonic, identifying the local audio signal as a voice signal.