Listener Preference for Wave Field Synthesis, Stereophony, and Different Mixes in Popular Music

Abstract

Stereophony and wave field synthesis are capable of providing the listener with a rich spatial audio experience. They both come with different advantages and challenges. A direct comparison in terms of listener preference has rarely been carried out in previous research, since these methods differ in their production stage. This study generated mixes of four different popular music recordings and introduced systematic changes to the wave field synthesis mix for one song. Listeners rated their preference comparing the different mixes and reproduction systems. Part of the experiment was repeated using a binaural simulation of the involved loudspeaker setups. The results show that a presentation using a high number of loudspeakers is preferred and the differences between reproduction methods can have a larger influence than strong variations of single mixing parameter. INTRODUCTIONThe most common loudspeaker reproduction methods in the consumer market are 2-or 2.1-channel stereophony (stereo) and 5.1-channel stereophony (surround / 5.1) [1]. Several multichannel methods that provide a higher number of surrounding loudspeakers are available as well [2]. More loudspeakers can lead to a higher emotional reaction and encourage deep immersion of listeners during reproduction [3]. The question arises if integrating more loudspeakers will also enhance the overall listening experience for listeners. In a study comparing stereo, surround, Ambisonics, 9-channel, and 22-channel stereophony Francombe et al. [4] showed that listeners preferred systems with more than five channels for different types of broadcast content.The present study assesses if the same preference for systems with a higher number of loudspeakers is present for wave field synthesis (WFS) [5,6] when compared to stereo and surround in the context of popular music. There is evidence against this as WFS is known to introduce stronger coloration on single sound sources than stereophony [7].Mixes of popular music for WFS are produced in a slightly different workflow than for stereo and surround. When listeners are asked which reproduction system they prefer for a given song the actual mix might influence their choice as well. The goal of the study is to quantify the influence of the mixing process on listener preference.The involved reproduction systems can be simulated via headphones, employing binaural re-synthesis to mimic the involved loudspeaker setups [8]. This is especially of interest for investigating loudspeaker-based methods that rely on a large number of loudspeakers, which might be hard to realize in a listening test [9]. The study employs binaural simulations of the loudspeaker reproduction systems and compares listener preferences obtained with those simulations and the actual loudspeaker setups.We assess those three questions in paired comparison listening tests asking listeners to rate their preference. Experiment I investigates the influence of the reproduction system and four different mixing parameter on listeners' preference ratings for one ...

A listener preference model for spatial sound reproduction, incorporating affective response

“…Some participants based their preference mostly on the content, while others on the system used for reproduction. Similar trends for OLE were also found from evaluations of 3D audio systems [4] and from comparisons of different reproduction systems [5,6]. In a previous work of the authors [21], a simple regression model for OLE prediction of binaurally recorded music excerpts in three different formats (mono, stereo and surround) was introduced and the results indicated that it is the preference for the content of the music excerpts that dominated on OLE formation, whilst a lesser impact was registered by the perceptual factors (in this case only Envelopment was considered) introduced by the audio system.…”

“…As with many other aspects of audio system performance, a growing body of recent research concerns methods that can also describe and predict the listener's experiences incorporating individual preferences for the content, i.e. the music material [4][5][6].…”