Effects on perception of auditory illusions

录音和回放系统的发展是为了创造一种完美的听觉错觉。在过去的几十年里，像双声音响、波场合成和双耳耳机回放这样的音频系统变得越来越重要。正在努力提高这类系统的质量。尽管经过多年的研究和发展，仍有许多知觉效应和心理声学因素尚不清楚或尚未完全了解。本项目系统分析双耳回放的感知效应。听力测试计划用来衡量质量的特征。在此基础上，建立了双耳合成场景感知的测量模型。双耳听化的合理性很大程度上取决于语境因素。为了测量这种影响，需要确定和评估一组质量特征。这些质量特征及其权重可以概括为感知的合理性。在项目期间，我们创建了几个不同的听觉场景来进行评估。影响特征质量的因素多种多样。不同的房间用于播放和录音，以及声学场景的复杂性，个性化程度和视觉呈现类型的变化。由于存在许多自由度，因此需要进行预测试和专家访谈，以确定最具影响力的因素。在进一步的听力测试中，测量了这些因素的感知效应。这些影响之间的依赖关系是预期的，并将被评估。在接下来的步骤中，对这些效果进行加权，以对应于它们对创造可信的听觉错觉的重要性。为了对每个质量特征进行大量的听力测试，需要开发一种特定的测试方法。

The development of recording and playback systems is motivated by the goal to create a perfect auditory illusion. In the last decades audio systems like ambisonics, wave field synthesis and binaural headphone playback gained in importance. Efforts are made to increase the quality of such systems. Despite the years of research and development, there are still many perceptual effects and psychoacoustical factors which are still unknown or not fully understood. This project deals with the systematical analysis of perceptual effects regarding binaural playback. Listening tests are planned to measure features of quality. Based on these findings a measurement model on the perception of binaural synthesized scenes is developed. The plausibility of a binaural auralisation significantly depends on contextual factors. To measure this influence, a set of quality features need to be determined and evaluated. These quality features and their weightings can be summarized as plausibility of perception. During the project several different auditive scenes are created for evaluation. The factors which influence the quality of features are varied. Different rooms for playback and recording are considered as well as variations of the complexity of the acoustical scene, the degree of personalization and the type of visual presentation. Because of the many degrees of freedom, pretests and interviews with experts are conducted to determine the most influential factors. In further listening test the perceptual effects of these factors are measured. Dependencies between these effects are expected and are going to be evaluated. In the next step, the effects are weighted corresponding to their importance for creating a plausible auditive illusion. To carry out the large amount of listening tests for each quality feature, a specific test methodology needs to be developed.