Objective evaluating method for audio quality of streaming media

Abstract

The invention discloses an objective evaluating method for the audio quality of streaming media, which comprises: acquiring original audio at a transmitter and acquiring distorted audios into which a coder, a decoder, package lost and delay jitter impairment are introduced after network transmission; performing pretreatment and module alignment treatment on distorted audios with a network characteristic, performing perceived evaluation of audio quality (PEAQ) and outputting the distorted audios of which delay and jitter are eliminated; performing coder and decoder and packet lost impairment quality evaluation on the original audio and aligned distorted audio; for impairment caused by delay and jitter between the distorted audios and the aligned audios, performing objective quality evaluation by network impairment estimation; and fitting the quality evaluation values of the two kinds of impairment to obtain the objective evaluation values of the original audios and the distorted audios transmitted by the network.

Description

technical field [0001] The invention relates to a communication network quality evaluation method, in particular to an objective evaluation method for streaming media audio quality. Background technique [0002] The 21st century is an era of rapid development of the network. With the popularization of the Internet, the demand for using the network to transmit audio signals is also increasing. The emergence of streaming media technology has improved the difficult situation of Internet audio transmission to a certain extent. It has changed the "push" communication of traditional media into "pull" communication and real-time communication of the audience. Since streaming media technology breaks through the limitation of network bandwidth on multimedia information transmission to a certain extent, it is widely used in various fields such as online live broadcast, network conference, distance education and enterprise training. In order to better transmit streaming media, it is u...

AI Technical Summary

Problems solved by technology

The advantage of subjective evaluation is that it is in line with people's perception of audio quality, but the disadvantage is that it is time-consuming, laborious and expensive, and it is not flexible enough, has poor repeatability and stability, and is greatly affected by people's subjective influence, etc.

[0004] Objective evaluation method In terms of speech: there is ITU-T P.862 (PESQ) suggestion, which is one of the objective evaluation algorithms for speech quality proposed by ITU. Improvements have been made, and it is currently the objective algorithm for voice quality evaluation with the highest correlation with the subjective score. The score is between 0 and 4.5. Normally, the output score is between 1.0 and 4.5, and the score is 2 points. Or less than 2 points, it means that the voice quality is relatively poor and difficult to understand, but it cannot give a good evaluation for the case of large audio and jitter delay; it has been widely used in VoIP ITU-T G. 107 specification (E model), its advantage is that it fully considers network damage factors such as noise, echo delay, encoder performance jitter, etc. in the network to give a single score R and has a considerable evaluation accuracy, and its quality level R value is the same as There is a non-linear and monotonous mapping relationship in the quality level of objective MOS scores, but the E-model has not been verified enough in the bounds and adequately measured in the laboratory for a large number of possible input parameter combinations, and the estimation is still Disadvantages such as being in the stage of doubt and research

In terms of audio: ITU-R synthesized six existing methods and some evaluation methods that are still under development and have great potential, and finally formed the ITU-RBS.1387 (PEAQ) recommendation, whose objective evaluation score is called ODG, Scores range from 0 to -4, with quality ratings: imperceptible (0 points), perceivable but not annoying (-1 points), slightly annoying (-2 points), annoying (-3 points), Very annoying (-4 points), however, as a proposed algorithm for audio codecs, there is a "training set problem", which can give good evaluations for high-speed and better-quality audio, however, for lower-quality audio If it is poor (such as low rate or high packet loss rate), it cannot give an evaluation with a high correlation with the subjective score

Since PEAQ is based on frame-to-frame comparison, for audio with delay jitter introduced, PEAQ cannot align the audio by itself and find the distorted frame corresponding to the original frame for comparison, so it cannot give a correct objective evaluation score

Images