CN-114464211-B - Cloud game audio quality testing method

CN114464211BCN 114464211 BCN114464211 BCN 114464211BCN-114464211-B

Abstract

The invention discloses a cloud game audio quality testing method which comprises the following steps that an audio testing end obtains test audio played on a testing client and a testing server in real time, forms first test audio and second test audio, obtains corresponding source audio in real time, obtains first test audio, audio file size value and audio track duration value of the source audio in the same time period respectively, calculates the ratio of the PESQ value, similarity value and audio file size value of the first test audio to the source audio, and compares the ratio of the audio size of the first test audio to the audio size of the source audio, the PESQ value and the similarity value with the ratio of the audio size of the source audio to the audio size of the source audio, the PESQ value and the similarity value of the source audio, which correspond to high-quality playing, standard-quality playing and fluent-quality playing of the source audio in audio playing software respectively. The invention can realize the automatic test of the cloud game audio quality, has high test speed, effectively improves the test efficiency and reduces the test cost.

Inventors

Zuo Wangwang
ZOU QIONG
ZHOU SHUANGQUAN

Assignees

深圳市瑞云科技有限公司

Dates

Publication Date: 20260505
Application Date: 20220216

Claims (4)

1. A method for testing the audio quality of a cloud game, comprising the steps of: the method comprises the following steps that S1, an audio test terminal controls a test server to play test audio, the test server transmits the test audio to a test client in real time, and the test audio is synchronously played through the test client; S2, the audio test terminal acquires test audio played on the test client terminal and the test server terminal in real time, and forms a first test audio and a second test audio respectively, and acquires corresponding source audio in real time; S3, the audio test terminal respectively acquires the audio file size value and the audio track duration value of the first test audio, the second test audio and the source audio corresponding to the same time period, and respectively calculates the PESQ value and the similarity value of the first test audio and the second test audio, the first test audio and the source audio and the second test audio and the source audio corresponding to the same time period; S4, the audio test terminal respectively judges whether the audio track duration values of the first test audio corresponding to the same time period are equal to the second test audio corresponding to the same time period, the first test audio corresponding to the same time period is equal to the source audio corresponding to the same time period, and respectively calculates the ratio of the audio file size values of the first test audio corresponding to the same time period to the second test audio corresponding to the same time period, the first test audio corresponding to the same time period is equal to the source audio corresponding to the same time period; S5, the audio test end respectively obtains the ratio of the audio sizes of the played audio and the source audio, the PESQ value and the similarity value corresponding to the high-quality playing, the standard-quality playing and the fluent-quality playing of the source audio in the audio playing software, compares the ratio of the audio sizes of the first tested audio and the source audio, the PESQ value and the similarity value with the ratio of the audio sizes of the played audio and the source audio corresponding to the high-quality playing, the standard-quality playing and the fluent-quality playing of the source audio in the audio playing software, the PESQ value and the similarity value, and outputs a comparison result; if the PESQ value of the first test audio and the second test audio or the first test audio and the source audio or the second test audio and the source audio calculated by the audio test end in the step S3 exceeds the PESQ threshold set on the audio test end, the output result of the audio test end is that the audio test end is adjusted to acquire the coding modes of the first test audio, the second test audio and the corresponding source audio in real time; If the similarity value between the first test audio and the second test audio or between the first test audio and the source audio or between the second test audio and the source audio calculated by the audio test end in the step S3 exceeds the similarity threshold set on the audio test end, the output result of the audio test end is that the tuning operation of the number of channels and the bit rate is required; If the audio test end determines in the step S4 that the audio track duration values of the first test audio and the second test audio corresponding to the same time period and/or the first test audio and the source audio and/or the second test audio and the source audio are not equal, the audio test end checks the instantaneity when the audio test end obtains the first test audio, the second test audio and the source audio in the same time period in the step S2.
2. The method for testing the audio quality of the cloud game according to claim 1, wherein communication connection is established among the testing server, the testing client and the audio testing terminal.
3. The method of claim 1, wherein the step S4 further comprises the step of the audio test terminal respectively judging whether the audio file sizes of the first test audio and the second test audio, the first test audio and the source audio, and the second test audio corresponding to the same time period are equal, and if the audio file sizes of the first test audio and the second test audio, the first test audio and the source audio, or the second test audio and the source audio corresponding to the same time period are not equal, the audio test terminal outputs the result that the transmission network and the transmission protocol between the test server terminal and the test client terminal, the test client terminal and the audio test terminal, and the test server terminal and the audio test terminal need to be checked.
4. The method for testing the audio quality of the cloud game according to claim 1, wherein the audio playing software is QQ music or internet cloud music or cool dog music.

Description

Cloud game audio quality testing method Technical Field The invention relates to the field of audio testing methods, in particular to a cloud game audio quality testing method. Background At present, a method special for cloud game audio quality test is not available in the market, in the prior art, evaluation and scoring are generally carried out after multi-person listening through manual listening, and the results of the multi-person evaluation and scoring are compared to realize the cloud game audio quality test. However, by comparing the feedback results after the multi-person listening, there is a comparison error, and by the manual listening, there is a certain subjectivity, and the accuracy is lower, finally the accuracy of the test is low, and at the same time, by the manual listening, the execution speed is slower, the time for comparing the data is long, and a lot of time is required to be input, resulting in low test efficiency. Accordingly, the prior art has drawbacks and needs improvement. Disclosure of Invention The invention aims to overcome the defects of the prior art, and provides a cloud game audio quality testing method which solves the problems of low testing accuracy and low testing efficiency caused by manually testing the cloud game audio quality in the prior art. The technical scheme of the invention is as follows, a cloud game audio quality testing method comprises the following steps: S1, an audio test terminal controls a test server terminal to play test audio, the test server terminal transmits the test audio to a test client terminal in real time, and the test audio is synchronously played through the test client terminal. And S2, the audio test terminal acquires test audios played on the test client terminal and the test server terminal in real time, forms a first test audio and a second test audio respectively, and acquires corresponding source audios in real time. And S3, the audio test terminal respectively acquires the audio file size value and the audio track duration value of the first test audio, the second test audio and the source audio corresponding to the same time period, and respectively calculates the PESQ value and the similarity value of the first test audio and the second test audio, the first test audio and the source audio and the second test audio and the source audio corresponding to the same time period. And S4, the audio test terminal respectively judges whether the audio track duration values of the first test audio corresponding to the same time period are equal to the second test audio corresponding to the same time period, the first test audio corresponding to the same time period is equal to the source audio corresponding to the same time period, and respectively calculates the ratio of the audio file size values of the first test audio corresponding to the same time period, the second test audio corresponding to the same time period, the first test audio corresponding to the same time period, and the second test audio corresponding to the same time period. S5, the audio test end respectively obtains the ratio of the audio sizes of the played audio and the source audio, the PESQ value and the similarity value corresponding to the high-quality playing, the standard-quality playing and the fluent-quality playing of the source audio in the audio playing software, compares the ratio of the audio sizes of the first tested audio and the source audio, the PESQ value and the similarity value with the ratio of the audio sizes of the played audio and the source audio, the peSQ value and the similarity value corresponding to the high-quality playing, the standard-quality playing and the fluent-quality playing of the source audio in the audio playing software, and outputs a comparison result. Further, communication connection is established among the test server, the test client and the audio test terminal. Further, if the PESQ value of the first test audio and the second test audio or the first test audio and the source audio or the second test audio and the source audio calculated by the audio test end in the step S3 exceeds the PESQ threshold set on the audio test end, the output result of the audio test end is that the audio test end is adjusted to obtain the encoding modes of the first test audio, the second test audio and the corresponding source audio in real time. Further, if the similarity value between the first test audio and the second test audio or between the first test audio and the source audio or between the second test audio and the source audio calculated by the audio test terminal in the step S3 exceeds the similarity threshold set on the audio test terminal, the audio test terminal outputs the result that the tuning operation of the number of channels and the bit rate is required. Further, if in the step S4, the audio test end determines that the first test audio corresponding to the same time period is not equal to the second test audio a