Statistical Inference of User Experience of Multichannel Audio on Mobile Phones

Mobile phones and other handheld electronic devices are now ubiquitous and play an important role in our everyday lives. Over the last decade, we have seen a sharp rise in the sophistication of both hardware and software for these devices, thus significantly increasing their utility and use. Electronic devices are now commonly used for the streaming of audio and video and for the regular playback of music. Multichannel audio has now become a popular format and with recent updates in software, the latest audio codecs that support this format can effectively be played back on most electronic devices. As a result, more audio content that is now compatible with these devices is being produced. It is important to perform subjective quality tests for multichannel audio verses traditional stereo and mono on devices like mobile phones while using headphones. This paper presents the results and analysis of such listening tests that evaluate basic audio quality and quality of experience. The results show that while multichannel audio scores higher than stereo and mono for both of these attributes, the difference in score was less for the ratings of quality of experience. The results were processed using an Analysis of Variance (ANOVA) to show the statistical significance. Calculation of the effect size showed that the audio content had little bearing on the results.