Fabian Brinkmann

Audibility and Interpolation of Head-Above-Torso Orientation in Binaural Technology

Head-related transfer functions (HRTFs) incorporate fundamental cues required for human spatial hearing and are often applied to auralize results obtained from room acoustic simulations. HRTFs are typically available for various directions of sound incidence and a fixed head-above-torso orientation (HATO). If-in interactive auralizations-HRTFs are exchanged according to the head rotations of a listener, the auralization result most often corresponds to a listener turning head and torso simultaneously, while-in reality-listeners usually turn their head independently above a fixed torso. In the present study, we show that accounting for HATO produces clearly audible differences, thereby suggesting the relevance of correct HATO when aiming at perceptually transparent binaural synthesis. Furthermore, we addressed the efficient representation of variable HATO in interactive acoustic simulations using spatial interpolation. Hereby, we evaluated two different approaches: interpolating between HRTFs with identical torso-to-source but different head-to-source orientations (head interpolation) and interpolating between HRTFs with the same head-to-source but different torso-to-source orientations (torso interpolation). Torso interpolation turned out to be more robust against increasing interpolation step width. In this case the median threshold of audibility for the head-above-torso resolution was about 25 degrees, whereas with head interpolation the threshold was about 10 degrees. Additionally, we tested a non-interpolation approach (nearest neighbor) as a suitable means for mobile applications with limited computational capacities.

Spectral equalization in binaural signals represented by order-truncated spherical harmonics

The synthesis of binaural signals from spherical microphone array recordings has been recently proposed. The limited spatial resolution of the reproduced signal due to order-limited reproduction has been previously investigated perceptually, showing spatial perception ramifications, such as poor source localization and limited externalization. Furthermore, this spatial order limitation also has a detrimental effect on the frequency content of the signal and its perceived timbre, due to the rapid roll-off at high frequencies. In this paper, the underlying causes of this spectral roll-off are described mathematically and investigated numerically. A digital filter that equalizes the frequency spectrum of a low spatial order signal is introduced and evaluated. A comprehensive listening test was conducted to study the influence of the filter on the perception of the reproduced sound. Results indicate that the suggested filter is beneficial for restoring the timbral composition of order-truncated binaural signals, while conserving, and even improving, some spatial properties of the signal.

On the authenticity of individual dynamic binaural synthesis

A simulation that is perceptually indistinguishable from the corresponding real sound field could be termed authentic. Using binaural technology, such a simulation would theoretically be achieved by reconstructing the sound pressure at a listeners ears. However, inevitable errors in the measurement, rendering, and reproduction introduce audible degradations, as it has been demonstrated in previous studies for anechoic environments and static binaural simulations (fixed head orientation). The current study investigated the authenticity of individual dynamic binaural simulations for three different acoustic environments (anechoic, dry, wet) using a highly sensitive listening test design. The results show that about half of the participants failed to reliably detect any differences for a speech stimulus, whereas all participants were able to do so for pulsed pink noise. Higher detection rates were observed in the anechoic condition, compared to the reverberant spaces, while the source position had no significant effect. It is concluded that the authenticity mainly depends on how comprehensive the spectral cues are provided by the audio content, and the amount of reverberation, whereas the source position plays a minor role. This is confirmed by a broad qualitative evaluation, suggesting that remaining differences mainly affect the tone color rather than the spatial, temporal or dynamical qualities.

Assessment of speech perception based on binaural room impulse responses

The two most important aspects in binaural speech perception — better-ear listening and spatial release from masking — can be modelled well with current prediction frameworks operating on binaural room impulse responses (BRIRs). To incorporate effects of reverberation, a model extension was recently proposed, splitting the BRIR into an early, useful and a late, detrimental part, before fed into the prediction framework. In a more recent work (Leclère et al., 2015) a relation between the applied splitting time, room properties and the resulting prediction accuracy was observed. This interaction was investigated here by measuring speech reception thresholds (SRTs) in quiet for four simulated rooms with systematically varied levels of reverberation and a constant room geometry. By linking the applied splitting time to room acoustic parameters, the mean prediction error with the binaural model by Jelfs et al. (2011) could be reduced by about 1 dB. Further, the prediction accuracy with pseudo-binaural signals, which can be captured with existing microphone arrays allowing for the evaluation of different head orientations in a post-processing step, was tested. Results were close to predictions with BRIRs, illustrating its suitability for practical assessment of binaural speech perception in existing rooms. All relevant data generated in the course is this work is publicly available from http://dx.doi.org/10.14279/depositonce-6725.

A Database of Anechoic Microphone Array Measurements of Musical Instruments

Violin (classical) Violin_historical 430 Hz G3-C7 G3-C7 Normal of the soundboard aiming at Mic#25, Scroll aiming at between Mic#02 and Mic#03 Daniel Deuter 200905-18 14:15 15:15 Instrument: Neil Kristóv Értz (1998, rev. 2003), Copy of Guaneri; Bow: Violin Bow (50 g), René Groppe (Metz, 1997), Copy of baroque model (17th century); Strings: E String: Catgut (Damian Dlugolecki, 0.62 mm); A String: Catgut (Nicholas Baldock, Kathedrale Strings, 0.74 mm); D String: Catgut (+0.25 mm Silver, 1.12mm, Nicholas Baldock, LuxLine); G String: Catgut, Silver, diameter unknown

First international round robin on auralization: Results of the acoustical evaluation

The round robin on auralization aimed at a systematic evaluation of room acoustic modeling software by means of comparing simulated and measured impulse responses. While a physical evaluation by means room acoustical parameters and spectro-temporal comparisons is addressed in an accompanying talk, here, we focus on an evaluation of perceptual differences arising in complex room acoustical scenarios. In these cases, a mere physical evaluation might not be able to predict the perceptual impact of the manifold interaction of different sound propagation phenomena in enclosed spaces such as reflection, scattering, diffraction, or modal behavior. For this purpose, dynamic auralizations of binaural room impulse responses that were simulated with different room acoustical modeling softwares were evaluated against their measured counterparts. For this purpose, listening tests were conducted using “plausibility” and “authenticity” as overall quality criteria and the Spatial Audio and Quality Inventory (SAQI) for a ...

First international round robin on auralization: Concept and database

For evaluating the performance of room acoustical simulations or numerical simulations in general, these are usually compared to corresponding measurements as a benchmark. Previous studies indicated that differences may result from neglecting wave effects (scattering, diffraction, attenuation at grazing incidence). However, it also proved to be a challenge to provide a precise representation of the primary and secondary structure (geometry, source and receiver characteristics, absorption and scattering coefficients) of the measured ground truth to be re-modeled in the simulation. The round robin on auralization aimed to overcome such shortcomings by generating a ground truth database of room acoustical environments provided to developers of room simulation software. The database includes a selection of acoustic scenes such as “single reflection,” or “coupled rooms” which isolate single acoustic phenomena, as well as three complex “real-world” environments of different size. Simulated monaural and binaural...