This research describes a method to auralize the effect of a three-dimensional directional pattern of an acoustic source in a reverberant environment using real acoustic measurements of laser sparks. Laser-induced breakdown (LIB) produces a massless and point-like acoustic source. The authors demonstrate the performance of a volumetric array of LIBs for synthesizing arbitrary radiation patterns, auralize the radiation pattern of a loudspeaker and compare the measured and synthesized impulse responses in a reverberant room, and evaluate the method using listening tests. The synthesized room response matched the target response both in room response reconstruction and in listening tests. The proposed method requires no previous knowledge about the room characteristics; auralization is performed by convolving a sound signal with the synthesized room impulse responses using the cloud of laser sparks. The quality of the synthesized version was rated to be excellent.
Download: PDF (HIGH Res) (1.3MB)
Download: PDF (LOW Res) (1.0MB)
This research addresses the challenge of incorporating dynamic range limiting of the magnitude response of HRTFs for hearing-impaired listeners who are using binaural hearing aids. In this application, a direct implementation of spatialization using the original HRTFs may pose a problem if some frequency areas are moved above the pain threshold or below the hearing threshold for hearing-impaired listeners. An extensive discrimination experiment was conducted with 38 normal-hearing participants to determine the maximum amount of limitation that could be applied before audible artifacts degraded the sound image. The experiment helped to determine the maximum limitation that can be applied so that at least half of the normal-hearing tested subjects cannot distinguish between the original and limited version of the spatial filters used. Although wide-dynamic range compression (WDRC) at the output of the hearing aids can produce benefits, pre-processing the HRTFs is preferable to limit its effect on the frequency spectrum and loudness. The objective is to optimize the HRTFs to diminish the risk that they are distorted by the WDRC processing in a crucial part of the sound spectrum.
Download: PDF (HIGH Res) (1.7MB)
Download: PDF (LOW Res) (275KB)
Adapting a stereo reproduction system to the listener’s position allows for reproduction of 2D object-based audio with better localization accuracy when the listener is located outside of the sweet spot. The adaptation is composed of two parts: a compensation system that updates the loudspeakers feeds so that these are delivered to the listener with an intended magnitude and phase independently of the listening position, and an object-based rendering system using conventional panning algorithms. Initial localization simulations using the velocity and energy localization vectors predicted that the frequency-dependent panning can pan virtual audio sources outside of the loudspeaker region at low frequency. A perturbation analysis showed that, in practice, localization of audio objects is robust when these are panned between the stereo region, and that the localization of objects outside of the stereo region is both sensitive to errors and affected by the accuracy of the video tracking system and the homogeneity of the radiation pattern of the different loudspeakers of the system. These same results were corroborated with subjective tests.
Download: PDF (HIGH Res) (3.6MB)
Download: PDF (LOW Res) (911KB)
Binaural synthesis enables headphone presentation of the same auditory impression that a listener would perceive if present in the original sound field. While the presentation of a binaural signal requires a head-related impulse response and/or a binaural room response, it also requires compensation for the headphone response. A method is proposed for automatically regularizing the inversion of a headphone transfer function for headphone equalization. The problem arises from the fact that the inversion cannot treat peaks and notches as being perceptually equivalent. Notch inversion can create high-Q resonances that can be very unpleasant. Evaluation of the proposed method indicates that it provides an inversion filter that can maintain the accuracy of the conventional regularized inverse method while limiting the inversion of notches in a perceptually acceptable manner. The results show that the proposed method can produce perceptually better equalization than the regularized inverse method used with a fixed regularization factor or the complex smoothing method used with a half-octave smoothing window.
Download: PDF (HIGH Res) (2.1MB)
Download: PDF (LOW Res) (383KB)
Open
Access
Audio engineers are increasingly looking to improve the spatial experience of surround sound by including loudspeakers in the vertical domain, so-called height channels. Listening tests were conducted to investigate the frequency dependency of localization thresholds on vertical interchannel crosstalk. The effect of frequency on localization was significant: thresholds were highest at low frequencies (-5.3 dB at 125 Hz and -3.03 dB at 250 Hz), falling to between -9 dB and -10.5 dB as the frequency increased beyond 1000 Hz. When configuring 3D microphone arrays, cardioid microphones would be a more appropriate choice for the height layer than omnidirectional microphones with regard to localizing source images near the main loudspeaker layer position. To better create a vertical phantom image, different localization thresholds could be applied to different frequency bands of the height channel signal. Control of localization thresholds can be achieved by manipulating the levels of single-octave bands within the height-channel signal rather than by manipulating the signal as a whole.
Download: PDF (HIGH Res) (2.5MB)
Download: PDF (LOW Res) (218KB)
Generating sufficient acoustical output with smaller loudspeakers requires maximum voice coil excursion, thereby exploiting the available mechanical resources of the motor and suspension. Digital preprocessing of audio input signals can be used to prevent a mechanical or thermal overload, which would generate excessive distortion and eventually damage the transducer. The most important characteristics of the protection algorithm are the measurement or adaptation error for the critical state variables, overshoot and undershoot of the activation control system, and latency added by the protection system. An inexpensive adaptive protection scheme can be realized by monitoring the voltage and current at the transducer terminals without using a mechanical sensor. The adaptive nonlinear control generates not only the largest AC amplitude and maximum output but also reduces the nonlinear distortion. The active linearization and stabilization of the transducer also needs a reliable mechanical protection system because overshooting the protection limits will cause an undesired increase of the nonlinear compensation signal, and that would require power amplifiers to provide a higher peak voltage.
Download: PDF (HIGH Res) (5.3MB)
Download: PDF (LOW Res) (509KB)
To quantify the quality of a loudspeaker system, previous research generally focused on the loudspeaker parameters or on specific aspects of the acoustic output. This research attempts to determine the perceived sound quality of a diverse set of stereo music fragments in a wide variety of environments without using source audio signal as a reference. Rather, the reference is created by making binaural recordings with a head and torso simulator using the best-quality loudspeakers in the ideal listening spot in the best-quality listening environment. The reproduced reference signal with the highest subjective quality is compared to the acoustic degraded loudspeaker output of the system being tested. This new perceptual modeling approach allows a direct comparison between the perceived quality of an excellent loudspeaker in a bad reproduction room/non-optimal listening spot with that of a poor loudspeaker in an excellent reproduction room/optimal listening spot using any musical fragment. The model shows a high average correlation (0.85) between objective and subjective measurements.
Download: PDF (HIGH Res) (1.9MB)
Download: PDF (LOW Res) (847KB)
Sound Field Control is a multifaceted topic that relates to any context involving the active management of audio within an acoustic environment. This can include the creation of independent sound zones, active control of noise, personal communication systems, electroacoustic manipulation of room acoustics, and replication of complex spatial sound fields using multichannel audio systems. Selected papers on sound zones from the recent SFC 2016 conference are summarized.
Download: PDF (797KB)
