Method and system for synthesizing audio data

Abstract

The invention relates to the field of data processing, more particularly to a method and system for synthesizing audio data. A client collects audio data and records a client clock corresponding to a collection time; according to a preliminary calibration value corresponding to the client, the client clock is calibrated to obtain a preliminary clock; the audio data and the preliminary clock are sent to a synthesis terminal; a period of time from calibration request sending by the client to calibration clock receiving is obtained at a preset time interval to form a transmission time set; the synthesis terminal calibrates the preliminary clock according to the minimum transmission time in the transmission time set to obtain a precise clock; and the synthesis terminal synthesizes audio data collected by more than two clients based on the precise clock. Therefore, a phenomenon echoing or wrong beating of the audio data synthesized by multi-path audio data can be avoided.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a method and system for synthesizing audio data. Background technique [0002] In the process of synthesizing multiple channels of audio data into one channel of audio data, it is necessary to calibrate the local clock of the client where each audio data resides with the clock of the server as a standard. However, the existing clock calibration methods often ignore the time error generated during the process of calibrating the clock between the client and the server. The reason for this time error is that the client wants to calibrate the value of the local clock at the first moment and initiates calibration to the server. Clock request. After receiving the request from the client to calibrate the clock, the server obtains the value of the current server clock as the standard value of the client’s local clock at the first moment after calibration. It should be noted that the value ...

AI Technical Summary

Problems solved by technology

However, the existing clock calibration methods often ignore the time error generated during the process of calibrating the clock between the client and the server. The reason for this time error is that the client wants to calibrate the value of the local clock at the first moment and initiates calibration to the server. Clock request. After receiving the request from the client to calibrate the clock, the server obtains the value of the current server clock as the standard value of the client’s local clock at the first moment after calibration. It should be noted that the value of the current server clock has been There is a difference in the actual standard value corresponding to the first moment of the local clock

Although the difference between the standard value of the client’s local clock calibration and the actual standard value may be very small, in a distributed environment, the difference can often reach a few seconds, resulting in disordered audio data beats or obvious echo

[0003] In a distributed environment, due to the difference between the clocks of the server and each client, it is impossible to accurately know the real time it takes for the client to send a calibration clock request to the server to the server to obtain the current clock, and it is also impossible to know the server The real time spent in the process of sending the standard value to the client, so that in a distributed environment, each client cannot guarantee to work on the same physical clock; in addition, the moment the client calibrates the clock will also be due to the device Factors such as the busyness of the network, the scheduling time difference, and the physical configuration of the device itself lead to differences in the clock memory of each device

Therefore, the accuracy of using the existing calibration clock method in a distributed environment is low, and it is easy to cause the rhythm of the synthesized audio data to be disordered or have obvious echoes

Images