Sunday, May 21, 15:00 — 18:00
EB05-01 Practical Method to Evaluate Noise Generation Systems
Roksana Kostyk (Presenting Author), Przemyslaw Maziewski (Author), Dominik Stanczak (Author)
The paper presents the method used to test the accuracy of a fully calibrated noise generation system. The method is based on a frequency response comparison of binaural recordings done in real and simulated environments. The paper will give examples coming from two different audio laboratories. It will also illustrate the influence of the head and torso unit’s miss-position on the noise reproduction accuracy.
Engineering Brief 324
EB05-02 Free Database of Low Frequency Corrected Head-Related Transfer Functions and Headphone Compensation Filters
Vera Erbes (Presenting Author), Matthias Geier (Author), Sascha Spors (Author), Hagen Wierstorf (Author)
A database of publicly available head-related transfer functions (HRTFs) of a KEMAR manikin together with headphone compensation filters for various headphone types is presented. The HRTFs are based on previously published data from Wierstorf et al. (2011) that have additionally been corrected for low frequencies. This compensates for missing information due to low excitation energy in this frequency range during the measurement and allows for shorter impulse responses. A further benefit is demonstrated by the interpolation of HRTFs via magnitude and phase that is only possible with consistent phase information. Both the low-frequency correction as well as the generation of the headphone compensation filters are accompanied by Matlab code to document the processing.
Engineering Brief 325
EB05-03 Dataset of In-the-Ear and Behind-the-Ear Binaural Room Impulse Responses Used for Spatial Listening with Hearing Implants
Florian Klein (Presenting Author), Anja Chilian (Author), Maria Gadyuchko (Author), Stephan Werner (Author)
The contribution presents a dataset of binaural room impulse responses (BRIRs) using a KEMAR head and torso simulator. Sixteen positions around the head are recorded in three rooms with differing room acoustics. The rooms represent a standardized listening lab, a room for rehabilitation of hearing diseases, and a large sized room. Additionally to the in-the-ear recordings, behind-the-ear BRIRs are recorded to simulate the microphone positions of hearing aid devices. The dataset is used in a research project to develop innovative methods and technologies for spatial listening and speech intelligibility using cochlear implants and bone conduction hearing aids. The dataset enables binaural resynthesis of different directions and rooms for research and rehabilitation.
Engineering Brief 326
EB05-04 Virtual Source Width in Binaural Synthesis with Frequency-Dependent Directions
Hengwei Su (Presenting Author), Toru Kamekawa (Author), Atsushi Marui (Author)
To control the perceived source width rendering by headphone, a method to distribute different frequency bands of a sound source to different directions by binaural synthesis was investigated. Three types of signals including two anechoic musical recordings and white noise were filtered and split into 1/3 octave bands, and each band was convolved with HRTFs from different directions within the intended source width range. Subjective listening tests were conducted to evaluate the performance of this process. There are no evident results showing that this method can successfully synthesis extended sound images. However, it suggests that the distribution of bands according to spectral characteristics of signals is necessary to synthesize sound image without displacement of localization.
Engineering Brief 327
EB05-05 Testing Babble Noise Reduction Performance of Headset Microphones
Antti Kelloniemi (Presenting Author), Ergo Esken (Author)
Babble noise is a typical and specific problem in open offices and call centers, which is why workers in these environments use headsets. Babble noise causes disturbance in these spaces, and it easily leaks through typical send noise suppression processing to far end in telecommunication. To improve the send direction signal-to-noise ratio, headsets are equipped with microphone booms with acoustic noise cancelling microphones or microphone arrays. A method to evaluate their efficiency in reducing babble noise is described in this paper.
Engineering Brief 328
EB05-06 Binaural Spatialization Methods for Indoor Navigation
Sylvain Ferrand (Presenting Author), François Alouges (Author), Matthieu Aussal (Author)
The visually impaired people are able to follow sound sources with a remarkable accuracy. They often use this ability to follow a guide in everyday activities or for practicing sports, like running or cycling. On the same principle, it is possible to guide people with spatialized sound. We have thus developed a navigation device to guide with sounds using binaural synthesis techniques. In this device we are using both localization information provided by a precise and low latency positioning system and heading data computed from an Inertial Measurement Unit. These positioning data are feeding an HRTF based binaural engine, producing spatialized sound in real-time and guiding the user along a way. The user follows the sound, quite naturally and without initial training. Experiments show that it is possible to guide a walker with enough precision.
Engineering Brief 329
EB05-07 The Two!Ears Database
Fiete Winter (Presenting Author), Alexander Raake (Author), Sascha Spors (Author), Hagen Wierstorf (Author)
TWO !EARS was an EU-funded project for binaural auditory modelling with ten international partners involved. Its main goal was to provide a computational framework for the modelling of active exploratory listening that assigns meaning to auditory scenes. As one outcome of the project, a database including data acquired by the involved partners as well as third-party measurements has been published. Among others, a large collection of Head Related Impulse Responses and Binaural Room Impulse Responses is part of the database. Further, results from psychoacoustic experiments conducted within TWO !EARS to validate the developed auditory model were added. For the usage of the database together with the TWO !EARS model, a software interface was developed to download the data from the database on demand.
Engineering Brief 330
EB05-08 Personalized HRTF Measurement and 3D Audio Rendering for AR/VR Headsets
Woon-Seng Gan (Presenting Author), Nitesh Kumar Chaudhary (Author), Nguyen Duy Hai (Author), JianJun He (Author), Santi Peksi (Author), Rishabh Ranjan (Author)
This e-Brief describes our recent work in acquiring a fast, personalized head related transfer function (HRTF) and a personalized 3D audio rendering headsets for augmented and virtual reality (AVR) headsets. Binaural signal acquisition and rendering are important tasks in capturing the idiosyncratic acoustics of the pinnae, head and torso, and playback via headphones to the left and right ears. We will highlight a personalized HRTF binaural acquisition cum 3D audio headphone playback system that can take advantage of our individual ear-head anthropometry information in 3D sound acquisition and rendering.
Engineering Brief 331
EB05-09 Effect of a Known Environment on the Estimation of Sound Source Distance
Shashank Aswathanarayana (Presenting Author)
The estimation of sound source distance has been a topic of research interest for a number of decades now. Humans are known to be good at localizing sound in the azimuth and elevation but are poor at estimating the sound source distance. This project looks at examining the effect of a known environment on the estimation of sound source distance. The project aims at initially testing the subjects perception of sound source in an unknown environment and then examining the effect of training the subject to the environment to see if training/learning the acoustics of the environment improves the estimation of the source distance.
Engineering Brief 332
EB05-10 The Effects of Decreasing the Magnitude of Elevation-Dependent Notches in HRTFs on Median Plane Localization
Jade Raine Clarke (Presenting Author), Hyunkook Lee (Author)
A binaural experiment was conducted to investigate whether a necessary magnitude of pinna related spectral notches in HRTFs exist. Individual HRIRs were measured at 0°, 30°, and 60° in the median plane for three subjects. The original HRTFs were manipulated so that dominant spectral notches between 5 and 10 kHz were filled in two different degrees. Localization tests were carried out with each subject judging each stimulus condition 15 times in a randomized order. It was found that for the 30° and 60° sources, two subjects tended to perceive the image to move upwards as pinna related notches were reduced. For 0°, however, an increase in front-back confusion occurred as a result of notch magnitude manipulation.
Engineering Brief 333
EB05-11 An Impulse Response Dataset for Dynamic Data-Based Auralization of Advanced Sound Systems
Chris Pike (Presenting Author), Michael Romanov (Author)
This engineering brief presents a freely-available binaural room impulse response (BRIR) dataset measured on a multichannel loudspeaker system. The 32-loudspeaker array includes all loudspeaker layouts specified in Recommendation ITU-R BS.2051. Measurements were carried out in an ITU-R BS.1116-compliant listening room using a Neumann KU100 dummy head microphone. BRIRs were measured at 2° steps of rotation of the dummy head. The dataset can be used for dynamic data-based auralization of multichannel loudspeaker signals, such as those generated by the so-called advanced sound systems described in ITU-R BS.2051, i.e., systems that can render surround sound with height signals from channel-based, object-based, and/or scene-based content representations. The dataset is made freely-available in the SOFA file format.
Engineering Brief 334
EB05-12 Equipment for Fast Measurement of Head-Related Transfer Functions
Jose J. Lopez (Presenting Author), Pablo Gutierrez-Parera (Author)
Binaural audio can become the future of spatial sound systems as more and more music is consumed on mobile devices through headphones. For a better experience, the binaural sound must be individualized for each subject through the use of their personal Head-Related Transfer Function (HRTF). The most straightforward way of personalization is to measure in-situ the HRTF. However, installations and set-ups for that purpose require anechoic chambers and complex motorized positioning systems. In this brief, we present an installation deployed in a non-anechoic room with multiple loudspeakers that provide a way of measuring the HRTF with an excellent resolution in the azimuthal plane and a sufficient resolution on elevation for common purposes.
Engineering Brief 335