Mixer and mixing method

Abstract

The invention discloses a mixer and a mixing method. The mixer comprises a framing unit, a voice detection unit being connected with the framing unit and being used for detecting whether all paths of signals after framing include voice signals, a time-varying filter connected with the voice detection unit, a volume calculating unit connected with the voice detection unit, a weight calculation unit connected with the volume calculating unit, a mixing unit connected with the time-varying filter and the weight calculation unit, and a post processing unit connected with the mixing unit. Therefore, the voice quality can be improved; and fairness of all participants is reflected.

Description

technical field [0001] The invention relates to a sound mixing technology, in particular to a sound mixer and a sound mixing method. Background technique [0002] Video conferencing and teleconferencing are forms of meetings held through communication networks, and they can provide real-time voice communication for participants in remote spaces. In order to obtain a close to real meeting communication atmosphere, sound mixing technology is indispensable, and the sound mixing technology will directly affect the voice quality of the meeting. Sound mixing technology is specifically divided into analog sound mixing technology and digital sound mixing technology, among which, digital sound mixing technology is widely used due to its advantages of high precision, large dynamic range and low noise. The basic principle of digital sound mixing technology is to superimpose the digital signals of each voice signal after analog-to-digital conversion to form a sound mixing output signal...

AI Technical Summary

Problems solved by technology

The weighted sound mixing technology disclosed in this existing document has the following problems: when each voice signal is mixed, the smaller the signal amplitude, the smaller the weight, which causes the small signal to become smaller after mixing, which is easy to It leads to large distortion; secondly, there are no more than 4 people speaking at the same time in a general video conference, and this method directly participates in the mixing without any processing of the line signal (containing noise) that is not speaking, which is easy to reduce the voice after mixing SNR

The automatic threshold mixing technology disclosed in this existing document has the following problems: since the signals judged to be "no speech state" do not participate in the sound mixing, there is no "no speech state" signal at all in the sound after mixing, making these "no speech state" signals Participants in the "speaking state" have no sense of presence; at the same time, participants will experience ups and downs from silence to speaking, which will affect their hearing; secondly, although this method ensures that the weight of small signals is ≥ 1, it still cannot guarantee the voice quality of small signals

[0010] It can be seen from the above that in the existing sound mixing technology, each signal (whether there is voice or not) directly participates in the sound mixing without any processing, or uses silence detection to reduce the number of signal paths participating in the sound mixing; Detection, unnecessary noise will be added in the mixing process, which will affect the voice quality

If silence detection is used to reduce the number of audio mixing channels, the participants in the "non-speaking state" will have no sense of participation; in addition, in the existing audio mixing technology, the amplitude of the voice signal is used as a standard to determine the audio frequency of each channel. The weight of the signal is superimposed again, but the loudness of the voice is not exactly equal to its amplitude, it depends on the amplitude and frequency of the voice signal

Images