Audio Engineering Society AES Paris 2016

AES Paris 2016
Engineering Brief EB3

EB3 - eBriefs 3: Lectures


Monday, June 6, 11:45 — 14:00 (Room 353)

Chair:
Christian Uhle, Fraunhofer Institute for Integrated Circuits IIS - Erlangen, Germany

EB3-1 The Aerodynamics Phenomena of a Particular Bass-Reflex PortVictor Manuel Garcia-Alcaide, Universitat Politècnica de Catalunya - Barcelona, Spain; Sergi Palleja-Cabre, Universitat Politècnica de Catalunya - Barcelona, Spain; R. Castilla, Universitat Politècnica de Catalunya - Barcelona, Spain; P. J. Gamez-Montero, Universitat Politècnica de Catalunya - Barcelona, Spain; Jordi Romeu, Universitat Politècnica de Catalunya - Barcelona, Spain; Teresa Pamies, Universitat Politècnica de Catalunya - Barcelona, Spain; Joan Amate, Amate Audio S.L. - Terrassa, Barcelona, Spain; Natalia Milan, Amate Audio S.L. - Barcelona, Spain
The aim of this paper is to study the aerodynamics phenomena of a particular bass-reflex port that causes unwanted noise in the audible frequency range. After discarding structural and mechanical vibration issues, the hypothesis that vortex shedding could be the source of the noise has been considered. Experimental and numerical evidences of the vortex, an analysis of its noise and the similarities between the real performance and the simulated one are presented. The simulations have been performed with axisymmetric geometries with the open source OpenFOAM toolbox. Additionally, three different experiments were carried out. First, acoustic signal experiments were done to analyze the response of the bass-reflex ports. Second, a mechanical vibration was tested in order to discard this source of noise. A good agreement has been found between numerical and experimental results, especially in the frequency band of the detected noise, around 1200 Hz. The presented CFD approach has proved a useful and cost-effective tool to face this kind of phenomena.
Engineering Brief 257 (Download now)

EB3-2 A Novel 32-Speakers Spherical SourceAngelo Farina, Università di Parma - Parma, Italy; Lorenzo Chiesi, University of Parma - Parma, Italy
The construction and test of a novel compact spherical source equipped with 32 individually driven 2" loudspeakers is presented. The new sound source is designed for making room acoustics measurements, emulating the directivity pattern of various music instruments or human talkers and singers. The 32 signals feeding the loudspeakers can be obtained by three different approaches: a set of High Order Ambisonics coefficients computed for emulating the polar pattern of a fixed directivity source a set of SPS (Spatial PCM Sampling) signals recorded around a real source, employing a corresponding set of 32 microphones placed on a sphere surrounding the real source, a matrix of FIR filters, designed employing a mathematical theory almost identical to the one developed for creating virtual microphones from a spherical microphone array [1]. The presentation will show details of the construction of the new loudspeaker array, and the results of the first tests performed for evaluating the capability of creating arbitrary polar radiation patterns.
Engineering Brief 258 (Download now)

EB3-3 Distracting NoiseThomas Sporer, Fraunhofer Institute for Digital Media Technology IDMT - Ilmenau, Germany; Tobias Clauß, Fraunhofer Institute for Digital Media Technology (IDMT) - Ilmenau, Germany; Nicolas Pachatz, Technical University of Ilmenau - Ilmenau, Germany; Clemens Müller, Technical University of Ilmenau - Ilmenau, Germany; Matthias-Fritz Melzer, Technical University of Ilmenau - Ilmenau, Germany; Judith Liebetrau, Fraunhofer IDMT - Ilmenau, Germany
Noise in domestic and work environments is usually measured based on noise power. This is not reflecting the fact that temporal and spectral structure of the noise, but also the activity of the test subject influences the annoyance. In addition there is a difference between artificial noise signals and noise signals that probably have a meaning to the listener. In this study 15 assessors evaluated the perception of 23 natural noise stimuli at four different levels in two different situations. The situations are spatial recordings of a library and a canteen. The test subjects are not focusing on listening but on tasks but told to indicate when noise is distorting their activities.
Engineering Brief 259 (Download now)

EB3-4 Noise-Robust Speech Emotion Recognition Using Denoising AutoencoderHun Kyu Ha, Gwangju Institute of Science and Technology (GIST) - Gwangju, Korea; Nam Kyun Kim, Gwangju Institute of Science and Technology (GIST) - Gwangju, Korea; Woo Kyeong Seong, Gwangju Institute of Science and Technology (GIST) - Gwangju, Korea; Hong Kook Kim, Gwangju Institute of Science and Tech (GIST) - Gwangju, Korea
In this paper, a method of noise-robust speech emotion recognition under music noises is proposed by using a denoising autoencoder (DAE) and a support vector machine (SVM). The proposed method first trains a DAE by using emotional speech signals corrupted by music noises. Then, the output values from a middle layer of the DAE are used as speech features. Next, an SVM is trained to classify emotions using the DAE features. The performance of the proposed method is compared with that of a conventional SVM classifier. Consequently, it is shown that the proposed method relatively improves the overall emotion recognition rate by 9.76% under music noise conditions, compared to the conventional method.
Engineering Brief 260 (Download now)

EB3-5 Non-Intrusive Rumble Filtering by VLF Crossfeed with High Filter SlopesDouglas Self, The Signal Transfer Company - London, UK
Vinyl discs create subsonic anti-phase signals because they are never perfectly flat and cause vertical stylus movement. This is often made worse by cartridge-arm resonance, giving amplitudes peaking around 10 Hz and requiring 40 dB of attenuation to reduce them to the vinyl noise floor. A conventional rumble filter needs very steep slopes to do this without unduly affecting the bottom of the audio band at 20 Hz. L-R crossfeed at low frequencies cancels the anti-phase signals, converting bass information to mono. This is not a new idea but has never caught on, probably because in published implementations the anti-phase filtering slope always comes out as –6dB/octave, no matter what order of lowpass filter is used to control the crossfeed. It is demonstrated that time-correction of the lowpass filter group delay with simple allpass filtering gives a much steeper slope of –18dB/octave for 2nd, 3rd, and 4th-order Butterworth filters, and intrusion into the audio band is minimized; this is believed novel. A practical design using 2nd-order filters was built and measured and gave the desired results.
Engineering Brief 261 (Download now)

EB3-6 The Misunderstood Transformer: “The Answer Lies in the Flux!”Michael Turner, Nidec Motor Corporation - Harrogate, UK
Whether used for power supply or signal interfacing, transformers are a key component of audio equipment. What is surprising is the extent to which these apparently simple devices are misunderstood. This engineering brief seeks to dispel some of the more common myths, to clarify the relationships between voltage, current, flux, and saturation and to thereby assist with proper design, selection, and application of (mainly) power transformers
Engineering Brief 262 (Download now)

EB3-7 Designing a Laboratory for Immersive ArtsChristopher Keyes, Hong Kong Baptist University - Kowloon, Hong Kong
This brief gives an overview of a facility dedicated to 3D sound and multi-screen video. It houses a control room and a theater with the region’s only 24.2 channel sound system and 5 permanent HD video screens. At roughly 200 m3 it is a relatively small facility but has many uses. In its construction we were afforded a wide range of possibilities for spatial configurations and equipment choice. It is hoped that presenting some detail on these design decisions, including choices available and ultimately implemented, may be of use for readers planning and budgeting their own facilities.
Engineering Brief 263 (Download now)

EB3-8 Design and Implementation of a Low-Latency, Lightweight, High-Performance Voice Interface Front-EndThierry Heeb, ISIN-SUPSI - Manno, Switzerland; Digimath - Sainte-Croix, Switzerland; Andrew Stanford-Jason, XMOS Ltd. - Bristol, UK; Tiziano Leidi, ISIN-SUPSI - Manno, Switzerland
Smart voice interfaces are the enabler of a new generation of consumer products such as network connected, voice enabled personal assistants. These are based on distributed architectures where voice is captured and pre-processed locally before being sent to remote servers for semantic analysis and response generation. A key element to achieve lowest cost and the best natural speech user experience is to keep latency to a minimum. This eBrief presents a lightweight, high-performance voice interface front-end software framework capable of handling multiple PDM microphones and integrating PDM to PCM conversion, high-resolution inter-channel delay, decimation, signal correction, and optional output data framing. The software forms a complete smart voice interface front-end running on the XMOS xCORE-200 architecture and achieving very low latency.
Engineering Brief 264 (Download now)

EB3-9 Multiphysical Simulation Methods for Loudspeakers—Nonlinear CAE-Based SimulationsAlfred Svobodnik, Konzept-X GmbH - Karlsruhe, Germany; Roger Shively, JJR Acoustics, LLC - Seattle, WA, USA; Marc-Olivier Chauveau, Moca Audio - Tours, France; Tommaso Nizzoli, Konzept-X GmbH - Karlsruhe, Germany
This is the third in a series of papers on the details of loudspeaker design using multiphysical computer aided engineering simulation methods. In this paper the simulation methodology for accurately modeling the nonlinear electromagnetics and structural dynamics of a loudspeaker will be presented. Primarily, the calculation of nonlinear force factor Bl(x), nonlinear inductance Le(x), and stiffness Kms(x) in the virtual world will be demonstrated. Finally, results will be presented correlating the simulated model results to the measured physical parameters. From that, the important aspects of the modeling that determine its accuracy will be discussed.
Engineering Brief 265 (Download now)


Return to Engineering Briefs

AES - Audio Engineering Society