SP-1 | Spatial Audio | R3 | 2018-11-14 | 12:30 - 14:30
moderated by Thomas Lund
Minimum BRIR grid resolution for interactive position changes in dynamic binaural synthesis
SP-1-1 | Start 12:30 | Duration 30 min. | Boris Reif | Talk (English)
This paper presents two psychoacoustic experiments on the minimum BRIR grid resolution required for interactive listener translation in virtual acoustic environments produced over headphones. A listening area with a size of 1 m x 2 m was set up with a positional resolution of up to 5 cm. 77490 BRIRs were simulated for this investigation.
A listening experiment was conducted to study the perception of discontinuities in the representation with different BRIR grid resolutions. 4 different test signals were taken into account. The results show, that the required resolution depends on the signal. For white noise, only the highest resolution (5 cm) brought the best results, while for dry male speech lower resolutions appeared to be just as fine.
Investigations on the Impact of Distance Cues in Virtual Acoustic Environments
SP-1-2 | Start 13:00 | Duration 30 min. | Christoph Pörschmann |
Johannes M. Arend, Philipp Stade | AES Reviewed Paper (English)
In this paper, we analyze different auditory distance cues in dynamic binaural synthesis. We compare the contributions of sound intensity, direct-to-reverberant ratio (DRR), and of near-field cues. For the auralization, we use the BinRIR method, which allows to generate binaural room impulse responses (BRIRs) for dynamic binaural synthesis based on a measured omnidirectional room impulse response (RIR). With BinRIR based on a simple geometric model, the listener position can be freely adjusted and the distance cues can be adapted separately. Furthermore, near-field head-related impulse responses (HRIRs) can be applied for direct sound and early reflections if the listener is very close to the virtual sound source. In a listening experiment, we presented stimuli at different distances in four synthesized rooms. In one condition, the stimuli contained natural distance-dependent intensity cues, and in another condition, the stimuli were loudness normalized. The results showed that even for loudness-normalized stimuli, an adequate distance perception can be obtained by adapting the DRR. The influence of near-field HRIRs, which were as well tested in the experiment, is very weak.
SP-1-3 | Start 13:30 | Duration 30 min. | Markus Zaunschirm |
Franz Zotter, Matthias Frank | Talk (English)
In the current effort to improve sound for virtual auditory environments with the upcoming affordable head mounted display solutions, realism and audio quality in head-tracked binaural rendering is again becoming important. While rendering based on static dummy-head measurements achieve high audio quality and externalization, the realism lacks interactivity with changes of the head orientation. Motion-tracked binaural (MTB) has been presented as a head-tracked rendering algorithm for recordings made with circular arrays on rigid spheres. In this contribution, we investigate the algorithm proposed for MTB rendering and adopt it for variable-orientation rendering using binaural room impulse responses (BRIR) measured for multiple, discrete orientations of a dummy head. Our listening experiment is based on such BRIR measurements at different loudspeaker distances. The experiment in particular investigates the perceptual implications of the angular resolution of the multi-orientation BRIR sets and the time/frequency-resolution of the algorithm.
Spatial vs. Static sound - the differences it makes for 360° videos, highlighted by heatmaps
SP-1-4 | Start 14:00 | Duration 30 min. | Martin Rieger |
Sylvia Rothe | Talk (English)
This talk reveals the results of a case-study he made with Sylvia Rothe (LMU Munich). The subject was to investigate the differences and similarities which head-tracked and head-locked stereo have on the viewer in terms of where a 360° video is being looked at with 50 subjects in total.
The purpose of the talk is to inform the listener what role spatial audio plays for 360° videos under real-life conditions with average viewers and if it is as important sound guys think it is.
Martin states background information about the topic and his motivation describes the experiment set-up and interprets the results of the heatmap as well as possible future development.
For measurement of the viewing direction, an app was created with Unity, which was then split into two apk-files. Both apps contained the same video, while one played the sound with ambisonics (.tbe, two big ears) and the other one represented a binaural stereo-downmix at 0° azimuth and 0° elevation. Each time the app was played, a 3-megabyte .vtt-files was written on the device, which were later collected. Therefore, Sylvia Rothe wrote a scientific paper on how to create such a heat map as part of her Ph.D. in virtual reality at LMU Munich.
Measured differences of user directions, acceleration, and subjective feeling will be featured as well as similarities that could be found in more both groups such as the degree of head-rotation and duration of hotspots.
Furthermore, Martin wants to motivate on doing similar work on VR because there is a lot more to find out.
PP-1 | Poster | Poster Zone | 2018-11-15 | 13:00 - 15:00
moderated by Michael Hösel
The following list of contributions (2 of 12 in total) to this event is incomplete due to filtering.
SOFAPan – Ein Binauralrenderer mit austauschbaren und individualisierbaren HRTF
PP-1-7 | David Bau |
Dieter Leckschat, Christian Epe | Poster (German)
Ein wichtiges Thema bei der Binauralen Synthese ist die einfache Verwendung personalisierter HRTF-Datensätze. An der Hochschule Düsseldorf wurde ein Binauralrenderer im DAW Plug In-Format entwickelt, der das noch recht unbekannte aber innovative SOFA Dateiformat nutzt, um verschiedene HRTF Datensätze schnell und einfach zu laden, zu analysieren und zu verwenden. Darüber hinaus ist es möglich, eine geladene HRTF auf den Hörer weiter anzupassen: Der individuelle Kopfdurchmesser eines Hörers beeinflusst maßgeblich die ITD (Interaural Time Difference) und hat damit einen maßgeblichen Einfluss auf die individuelle Lokalisationsfähigkeit. Im Renderer wurde ein spezielles Verfahren implementiert, um eine geladene HRTF auf den individuellen Kopfdurchmesser des Hörers einzustellen.
Mit der implementierten Nahfeldsimulation ist es außerdem möglich, auch mit einem im Fernfeld gemessenen HRTF-Datensatz die auftretenden Effekte einer in unmittelbarer Nähe des Kopfes befindlichen Schallquelle zu simulieren.
A position-dynamic binaural synthesis of a multi-channel loudspeaker setup as an example of an auditory augmented reality application
PP-1-8 | Stephan Werner |
Mina Voigt | Poster (English)
This contribution presents an insight into a position-dynamic binaural synthesis system used in an auditory augmented reality (AR) application. Audio signals from six loudspeakers are reproduced binaurally in a real room. The listener can move freely within an area of approx. 4 m x 4 m surrounded by a virtual loudspeaker setup. The head pose and the listener position are tracked and the corresponding binaural room impulse responses are continuously provided. A setup of five loudspeakers corresponds to a surround setup. One additional loudspeaker is located within the walkable area. A self-created music installation is reproduced with the virtual loudspeaker setup. This study investigates techniques and concepts for creating an interactive musical experience with such a system. Technically, panned audio signals between the surround speakers and discrete audio objects at the several speakers are used to create an acoustic scene. The listener's movements are incorporated as a design element.
PP-2 | Poster | Poster Zone | 2018-11-16 | 13:00 - 15:00
moderated by Michael Hösel
The following list of contributions (2 of 12 in total) to this event is incomplete due to filtering.
SOFAPan – Ein Binauralrenderer mit austauschbaren und individualisierbaren HRTF
PP-2-7 | David Bau |
Dieter Leckschat, Christian Epe | Poster (German)
Ein wichtiges Thema bei der Binauralen Synthese ist die einfache Verwendung personalisierter HRTF-Datensätze. An der Hochschule Düsseldorf wurde ein Binauralrenderer im DAW Plug In-Format entwickelt, der das noch recht unbekannte aber innovative SOFA Dateiformat nutzt, um verschiedene HRTF Datensätze schnell und einfach zu laden, zu analysieren und zu verwenden. Darüber hinaus ist es möglich, eine geladene HRTF auf den Hörer weiter anzupassen: Der individuelle Kopfdurchmesser eines Hörers beeinflusst maßgeblich die ITD (Interaural Time Difference) und hat damit einen maßgeblichen Einfluss auf die individuelle Lokalisationsfähigkeit. Im Renderer wurde ein spezielles Verfahren implementiert, um eine geladene HRTF auf den individuellen Kopfdurchmesser des Hörers einzustellen.
Mit der implementierten Nahfeldsimulation ist es außerdem möglich, auch mit einem im Fernfeld gemessenen HRTF-Datensatz die auftretenden Effekte einer in unmittelbarer Nähe des Kopfes befindlichen Schallquelle zu simulieren.
A position-dynamic binaural synthesis of a multi-channel loudspeaker setup as an example of an auditory augmented reality application
PP-2-8 | Stephan Werner |
Mina Voigt | Poster (English)
This contribution presents an insight into a position-dynamic binaural synthesis system used in an auditory augmented reality (AR) application. Audio signals from six loudspeakers are reproduced binaurally in a real room. The listener can move freely within an area of approx. 4 m x 4 m surrounded by a virtual loudspeaker setup. The head pose and the listener position are tracked and the corresponding binaural room impulse responses are continuously provided. A setup of five loudspeakers corresponds to a surround setup. One additional loudspeaker is located within the walkable area. A self-created music installation is reproduced with the virtual loudspeaker setup. This study investigates techniques and concepts for creating an interactive musical experience with such a system. Technically, panned audio signals between the surround speakers and discrete audio objects at the several speakers are used to create an acoustic scene. The listener's movements are incorporated as a design element.