Figure 1 - uploaded by Maarten Van Walstijn
Content may be subject to copyright.
The 10 loudspeaker positions used for the ambisonic arrays, the targets and the three listening positions. The eight loudspeakers used for the third order HOA array are circles and the four that were also used for the first order array are coloured black. Those used as targets are indicated by red crosses. The listening positions are shown as blue shapes around the centre.
Source publication
Ambisonics is a scalable spatial audio technique that attempts to present a sound scene to listeners over as large an area as possi-ble. A localisation experiment was carried out to investigate the performance of a first and third order system at three listening positions -one in the centre and two off-centre. The test used a reverse target-pointer...
Contexts in source publication
Context 1
... lis- tening positions were at (0, 0) m, (0, −0.6) m and (−0.42, 0.52) m, where the origin is the centre of the loudspeaker array, the x- axis is to the front and y-axis is to the left. The listening positions are called centre, right and back-left respectively throughout this paper and are indicated in figure 1. ...
Context 2
... array had a radius of 2.2m and consisted of eight equalised Genelec 1030a loudspeakers. figure 1. All eight were used for the o3spk8 system. ...
Similar publications
n this paper, we propose a new scheme for audio steganography in the spatial domain using a transparent LSB matching approach and then we test the performance of the scheme with standard LSB substitution algorithms and well-known audio steganography tools. By using a data hiding approach similar to image LSB matching, it has been observed that the...
Compensated Amplitude Panning (CAP) is a spatial audio reproduction method for loudspeakers that takes the listener head orientation into account. Using CAP it is possible to produce stable images in all directions using only two loudspeakers. In its original formulation CAP is inherently an object-based method, with each image produced separately....
Recent text-to-audio generation techniques have the potential to allow novice users to freely generate music audio. Even if they do not have musical knowledge, such as about chord progressions and instruments, users can try various text prompts to generate audio. However, compared to the image domain, gaining a clear understanding of the space of p...
Recent text-to-audio generation techniques have the potential to allow novice users to freely generate music audio. Even if they do not have musical knowledge, such as about chord progressions and instruments, users can try various text prompts to generate audio. However, compared to the image domain, gaining a clear understanding of the space of p...
Ambisonics and Higher Order Ambisonics (HOA) are scalable spatial audio techniques that attempt to present a sound scene to listeners over as large an area as possible. A localisation experiment was carried out to investigate the performance of a first and third order system at three
listening positions - one in the centre and two off-centre - usin...
Citations
... All three methods were deemed suitable to evaluate hearing aid performance, but different algorithms and metrics required different optimal solutions. Frank et al. (2008) and Stitt et al. (2013Stitt et al. ( , 2014 studied localization ability in the loudspeaker array center and for off-center positions. Frank et al. (2008) found similar performance for the HOA basic and HOA max r E methods in the center and slightly better results for the HOA max r E method at the off-center position. ...
... Frank et al. (2008) found similar performance for the HOA basic and HOA max r E methods in the center and slightly better results for the HOA max r E method at the off-center position. Stitt et al. (2013Stitt et al. ( , 2014 verified that increasing the order of the Ambisonics reproduction method improved localization. ...
The use of virtual acoustic environments has become a key element in psychoacoustic and audiologic research, as loudspeaker-based reproduction offers many advantages over headphones. However, sound field synthesis methods have mostly been evaluated numerically or perceptually in the center, yielding little insight into the achievable accuracy of the reproduced sound field over a wider reproduction area with loudspeakers in a physical, laboratory-standard reproduction setup. Deviations from the ideal free-field and point-source concepts, such as non-ideal frequency response, non-omnidirectional directivity, acoustic reflections, and diffraction on the necessary hardware, impact the generated sound field. We evaluate reproduction accuracy in a 61-loudspeaker setup, the Simulated Open Field Environment, installed in an anechoic chamber. A first measurement following the ISO 8253-2:2009 standard for free-field audiology shows that the required accuracy is reached with critical-band-wide noise. A second measurement characterizes the sound pressure reproduced with the higher-order Ambisonics basic decoder, with and without max rE weighting, vector base amplitude panning, and nearest loudspeaker mapping on a 187 cm × 187 cm reproduction area. We show that the sweet-spot size observed in measured sound fields follows the rule kr≤N/2 rather than kr≤N but is still large enough to avoid compromising psychoacoustic experiments.
... With its increased spatial resolution, HOA achieves better localization of the reproduced sound events on surrounding loudspeaker systems [10,11]. Especially at off-center listening positions, the improvement is substantial [12][13][14][15][16], resulting in an increased sweet area [17]. Because of the decreased inter-channel crosstalk, decorrelation of signals can be largely preserved in playback, improving the perception of spatial depth [18]. ...
For the creation of convincing virtual acoustics of existing rooms and spaces, it is useful to apply measured Ambisonic room impulse responses (ARIRs) as a convolution reverb. Typically, tetrahedral arrays offering only first-order resolution are the preferred practical choice for measurements, because they are easily available and processed. In contrast, higher order is preferred in playback because it is superior in terms of localization accuracy and spatial clarity. There are a number of algorithms that enhance the spatial resolution of firstorder ARIRs. However, these algorithms may introduce coloration and artifacts. This paper presents an improvement of the Ambisonic Spatial Decomposition Method by using four directions simultaneously. The additional signals increase the echo density and thereby better preserve the diffuse sound field components during the process of enhancing measured firstorder ARIRs to higher orders. An instrumental validation and a series of listening experiments compare the proposed Four-Directional Ambisonic Spatial Decomposition Method to other existing algorithms and prove its similarity to the best algorithm in terms of enhanced spatial clarity and coloration while producing the least artifacts.
... However, in the current experiment participants were seated in the center of the array. For offcenter listenening positions, studies have found that increasing Ambisonics order improves localization accuracy (Stitt et al., 2013, Stitt et al., 2014 also beyond the third order (Frank et al., 2008). Moreover, the current study tested localization in anechoic conditions. ...
To reproduce realistic audio-visual scenarios in the laboratory, Ambisonics is often used to reproduce a sound field over loudspeakers and virtual reality (VR) glasses are used to present visual information. Both technologies have been shown to be suitable for research. However, the combination of both technologies, Ambisonics and VR glasses, might affect the spatial cues for auditory localization and thus, the localization percept. Here, we investigated how VR glasses affect the localization of virtual sound sources on the horizontal plane produced using either 1st-, 3rd-, 5th- or 11th-order Ambisonics with and without visual information. Results showed that with 1st-order Ambisonics the localization error is larger than with the higher orders, while the differences across the higher orders were small. The physical presence of the VR glasses without visual information increased the perceived lateralization of the auditory stimuli by on average about 2°, especially in the right hemisphere. Presenting visual information about the environment and potential sound sources did reduce this HMD-induced shift, however it could not fully compensate for it. While the localization performance itself was affected by the Ambisonics order, there was no interaction between the Ambisonics order and the effect of the HMD. Thus, the presence of VR glasses can alter acoustic localization when using Ambisonics sound reproduction, but visual information can compensate for most of the effects. As such, most use cases for VR will be unaffected by these shifts in the perceived location of the auditory stimuli.
... However, in the current experiment participants were seated in the center of the array. For offcenter listenening positions, studies have found that increasing Ambisonics order improves localization accuracy (Stitt et al., 2013, Stitt et al., 2014 also beyond the third order (Frank et al., 2008). Moreover, the current study tested localization in anechoic conditions. ...
No PDF available
ABSTRACT
To reproduce realistic audio-visual scenarios in the laboratory, Ambisonics is often used to present a sound field over loudspeakers and virtual reality glasses are used to present visual information. However, the application of both technologies, Ambisonics and virtual reality glasses, might affect the spatial cues for auditory localization, resulting in reduction of the localization accuracy. Furthermore, the combination of both technologies might introduce further errors. Here, we investigated how a head-mounted display affects the localization of virtual sound sources produced using either 1st, 3rd, 5th or 11th order Ambisonics with and without visual information. Preliminary results suggest that there is an effect of Ambisonics order on localization accuracy, mostly in the first order Ambisonics, but that the virtual reality glasses do not add an additional error. The localization error with first order ambisonics remained even with visual information as the perceived location matched better with other visual sources (i.e., nearby loudspeakers). Thus, virtual reality can be used in combination with Ambisonics without additional localization errors, but for localization accuracy it is better to use higher order Ambisonics.
... How can we efficiently determine the perceptual sweet area? Typically, the sweet area is determined by assessing one or more quality features at various listening positions within the playback volume [11,12,13,14,15,16,17]. A comparison to the results of a reference decides which listening positions fulfill the requirements for the sweet area. ...
... An extension to the Gerzon energy vector was presented in [9] that improved the prediction accuracy for two localization experiments for off-center listening positions with first-and third-order Ambisonics [10,11]. The extension is a method for calculating perceptual weights that are applied to the loudspeaker gains before calculating the final vector. ...
... The two listening positions were the same as those used in [10] and [11], except the center seat was excluded here. These were one seat to the right and one diagonally to the back-left. ...
... The non-transient sound was 20 Hz to 20 kHz white noise with a linear amplitude increase from zero throughout the entire signal duration (206 ms), to avoid a strong onset. The target signal, played through only one loudspeaker, was amplitude-modulated white noise with a frequency range of 20 Hz to 20 kHz, as used in the experiments in [10,11]. ...
In multichannel reproduction, listening at off-centre positions involves differences in loudspeaker signal arrival times, meaning that the precedence effect is an important factor in localisation. Given that the type of signal is known to affect elements of the precedence effect, there is a need to understand the implications regarding the perceived auditory source image angle and the prediction thereof. In this paper the results of a localisation experiment for off-centre listening positions are presented, in which subjects were tested for transient and non-transient noise signals using third-order Ambisonics. These results are compared to those obtained with several prediction models, including an extended version of the energy vector model that incorporates the precedence effect. Compared to two binaural models, the extended vector model is shown to provide the best predictions over all conditions. The results confirm that the type of signal must be taken into account in predictive modelling. Furthermore, the extended energy vector exhibits about 50% less error than the standard energy vector.
... These models give good prediction for a centred listener, for which the phantom source direction is determined entirely by the loudspeaker gains. This does not hold for off-centre listening though, in which case unwanted arrival time delays (ATDs) between loudspeakers are introduced, which can trigger the precedence effect [12,13,14]. This is the suppression of later arriving copies (lag) of a lead sound (generally reflections in the case of real sound sources) to allow the direction of the sound source to be perceived without the corrupting influence of the reflections on the localisation cues. ...
... In section 3, it is necessary to mimic the target-pointer experiment task in [14,37], where an Ambisonic sound (pointer) was moved by the listener to match the direction of a single loudspeaker (target). Therefore, the lookup tables are generated for each of the Ambisonic system and listening position combinations with a resolution of 1 • Ambisonic panning angles. ...
... The two binaural models outlined in section 2, the standard energy vector and the extended energy vector, are compared to the median results of two previously published experiments using a 2.2 m radius loudspeaker array [14] and a 5 m radius [37]. First and third-order Ambisonics systems were tested for both array sizes and the max r E weighting was used in all cases. ...
accompanying webpage: http://www.socasites.qub.ac.uk/mvanwalstijn/jaes16/
This paper presents an extension to the energy vector, well known in the Ambisonics literature,
to improve its predictions of localization at off-center listening positions. In determining the source direction, a perceptual weight is assigned to each loudspeaker gain, taking into account the relative arrival times, levels, and directions of the loudspeaker signals. The proposed model is evaluated alongside the original energy vector and two binaural models through comparison with the results of recent perceptual studies. The extended version was found to provide results that were at least 50% more accurate than the second best predictor for two experiments involving off-center listeners with first- and third-order Ambisonics systems.
... The results of this experiment were presented at DAFx-2013 in Maynooth, Ireland in [140]. This section presents an extended version of these results. ...
... The experiment was carried out on a 5 m radius loudspeaker array and the results are compared to earlier ones obtained from a 2.2 m radius array. The first order system had higher overall error than the third order one, as shown in previous localisation experiments [4,15,51,140]. The pointer was shifted towards the further side of the loudspeaker array to compensate for the proximity of the nearest loudspeakers, which have both earlier arrival times and increased gains relative to the centre position. ...
Higher Order Ambisonics is a spatial audio technique that aims to recreate a sound image over as large a listening area as possible. Only limited investigation has taken place into localisation with Ambisonics and Higher Order Ambisonics at off-centre listening positions. This thesis presents the results of three psychoacoustic localisation experiments investigating off-centre localisation of first and third order Ambisonics under different conditions: in acoustically damped studio conditions, investigating transient versus ongoing, non-transient stimuli for the third order system, and for a large array system with increased arrival time delay between loudspeakers. A detailed analysis of the results of each experiment is carried out to determine the robustness of the tested ambisonic systems and whether the variation is principally due to differences between listeners or through the variation of individual listeners. Comparisons are made between the results of the three experiments to determine the influence of changing the stimulus or increasing the arrival time delay between loudspeakers, where the relative gains of the loudspeakers was found to be perceptually more important than increased time differences between them.
The significance and usefulness of these results can be increased by comparison with models for prediction of human localisation, where a robust model would afford fast evaluation of ambisonic systems and allow system optimisation for off-centre positions. Therefore, evaluation is performed for two binaural models on their ability to predict the results of the psychoacoustic experiments. This includes a model that has previously been used for prediction of off-centre listening positions for higher order systems and WFS. A modified version of the Lindemann model, which includes a precedence effect inhibition, is also evaluated, having previously been applied to two channel lead-lag experiments. Finally, the energy vector model, which is prevalent in the Ambisonics community, is extended to include elements of the precedence effect. The binaural models, the standard energy vector, previously untested for off-centre positions, and the precedence-extended model are evaluated by comparison to the perceptual results. The robustness of each of the binaural and vector models is discussed in the context of their use as predictors of localisation at off-centre listening positions. The predictions of the precedence-extended energy vector are found to exhibit the lowest deviation from the perceptual results.
... An experiment presented in an earlier paper by the authors [13] was specifically designed to investigate the localisation performance of ambisonic reproduction of two different orders (first and third) at three listening positions, including the centre as reference and the array had a radius of 2.2 m. It was found that the ambisonic image was moved to the further side of the array to compensate for the proximity to the nearest loudspeakers. ...
... The point at which this happens is known as the echo threshold. The listening experiment in [13] was repeated for an array radius of 5 m in order to test the influence of increased time delays on the localisation of ambisonic images. The listeners were the same proportional distance from the centre as in the first experiment so the relative loudspeaker gains would be approximately equal in both experiments and the time difference between different loudspeaker signal arrivals would increase. ...
... The listeners were the same proportional distance from the centre as in the first experiment so the relative loudspeaker gains would be approximately equal in both experiments and the time difference between different loudspeaker signal arrivals would increase. In [13] the arrival time delays had a range up to 4 ms (beyond the 1 ms localisation dominance threshold) and yet it was still possible to localise the ambisonic image when listening at the off-centre positions. For the 5 m array the range increases to approximately 9 ms (given in table IV) so it might be expected that the increased time delays mean the later signals will contribute less to the localisation of the image, increasing the difficulty and impacting on the result. ...
Ambisonics and Higher Order Ambisonics (HOA) are scalable spatial audio techniques that attempt to present a sound scene to listeners over as large an area as possible. A localisation experiment was carried out to investigate the performance of a first and third order system at three
listening positions - one in the centre and two off-centre - using a 5 m radius loudspeaker array. The results are briefly presented and compared to those of an earlier experiment on a 2.2 m loudspeaker array. In both experiments the off-centre listeners were placed such that the ratio of
distance from the centre to the array radius was constant in both experiments. The test used a reverse target-pointer adjustment method to determine the error, both signed and absolute, for each combination of listening position and system. The results for both arrays were found to be very
similar, suggesting that the relative amplitude of the loudspeakers, which were the same in both cases, was more dominant for localisation than the arrival time differences, which differed between array sizes.
... A recent study [26,27] investigated the inuence of the loudspeaker array radius on the o-center localization errors. For two dierent radii, the errors at the same relative positions were similar, cf. Figure 4. ...
... Medians and corresponding interquartile ranges of horizontal localization errors in dependence of listening position and Ambisonics order for 2 loudspeaker array radii[26,27]. ...
