AES Papers: AES 125th Convention

7501
A Parametric Instrument Codec for Very Low Bit Rates
Arnold, Mirko; Schuller, Gerald
A technique for the compression of guitar signals is presented which utilizes a simple model of the guitar. The goal for the codec is to obtain acceptable quality at significantly lower bitrates compared to universal audio codecs. This instrument codec achieves its data compression by submitting an excitation function and model parameters to the receiver instead of the waveform. The parameters are extracted from the signal using weighted least squares approximation in the frequency domain. For evaluation a listening test has been conducted and the results are presented. They show that this compression technique provides a quality level comparable to recent universal audio codecs. The usability however is at this stage limited to very simple guitar melody lines.

7502
Stereo ACC Real-Time Audio Communication
Ferreira, Anibal; Abreu, Filipe; Sinha, Deepen
Audio Communication Coder (ACC) is a codec that has been optimized for monophonic encoding of mixed speech/audio material while minimizing codec delay and improving intrinsic error robustness. In this paper we describe two major recent algorithmic improvements to ACC: on-the-fly bit rate switching and coding of stereo. A combination of source, parametric and perceptual coding techniques allows a very graceful switching between different bit rates with minimal impact on the subjective quality. A real-time GUI demonstration platform is available that illustrates the ACC operation from 16 kbit/s mono till 256 kbit/s stereo. A real-time two-way stereo communication platform over Bluetooth has also been implemented that illustrates the ACC operational flexibility and robustness in error-prone environments.

7503
MPEG-4 Enhanced Low Delay AAC - A New Standard for High Quality Communication
Schnell, Markus; Schmidt, Markus; Jander, Manuel; Albert, Tobias; Geiger, Ralf; Ruoppila, Vesa; Ekstrand, Per; Bernhard, Grill
The MPEG Audio group has recently concluded the standardization process for the MPEG-4 Enhanced Low Delay AAC (AAC-ELD) codec. This codec is a new member of the MPEG Advanced Audio Coding family. It represents the efficient combination of the AAC Low Delay codec and the Spectral Band Replication (SBR) technique known from HE-AAC. This paper provides a complete overview of the underlying technology, presents points of operation as well as applications and discusses MPEG verification test results.

7504
Efficient Detection of Exact Redundancies in Audio Signals
Zapata G., José; Garcia, Ricardo A.
An efficient method to identify bitwise identical long-time redundant segments in audio signals is presented. It uses audio segmentation with simple time domain features to identify long term candidates for similar segments, and low level sample accurate metrics for the final matching. Applications in compression (lossy and lossless) of music signals (monophonic and multichannel) are discussed.

7505
An Improved Distortion Measure for Audio Coding and a Corresponding Two-Layered Trellis Approach for Its Optimization
Melkote, Vinay; Rose, Kenneth
The efficacy of rate-distortion optimization in audio coding is constrained by the quality of the distortion measure. The proposed approach is motivated by the observation that the Noise-to-Mask Ratio (NMR) measure, as it is widely used, is only well adapted to evaluate relative distortion of audio bands of equal width on the Bark scale. We propose a modification of the distortion measure to explicitly account for Bark bandwidth differences across audio coding bands. Substantial subjective gains are observed when this modification over plain NMR is used in the Two Loop Search, for quantization and coding parameters of scalefactor bands in an AAC encoder. Comprehensive optimization of the new measure, over the entire audio file, is then performed using a two-layered trellis approach, and yields nearly artifact-free audio even at low bit-rates.

7507
Spatial Audio Scene Coding
Goodwin, Michael; Jot, Jean-Marc
This paper provides an overview of a framework for generalized multichannel audio processing. In this Spatial Audio Scene Coding (SASC) framework, the central idea is to represent an input audio scene in a way that is independent of any assumed or intended reproduction format. This format-agnostic parameterization enables optimal reproduction over any given playback system as well as flexible scene modification. The signal analysis and synthesis tools needed for SASC are described, including a presentation of new approaches for multichannel primary-ambient decomposition. Applications of SASC to spatial audio coding, upmix, phase-amplitude matrix decoding, multichannel format conversion, and binaural reproduction are discussed.

7508
Microphone Front-Ends for Spatial Audio Coders
Faller, Christof
Spatial audio coders, such as MPEG Surround, have enabled low bitrate and stereo backwards compatible coding of multi-channel surround audio. Directional audio coding (DirAC) can be viewed as spatial audio coding designed around specific microphone front-ends. DirAC is based on B-format spatial sound analysis and has no direct stereo backwards compatibility. We are presenting a number of two capsule based stereo compatible microphone front-ends and corresponding spatial audio coder modifications which enable the use of spatial audio coders to directly capture and code surround sound.

7509
Spatialized Additive Synthesis of Environmental Sounds
Verron, Charles; Aramaki, Mitsuko; Kronland-Martinet, Richard; Pallone, Grégory
In virtual auditory environment, sound sources are typically created in two stages: the dry monophonic signal is synthesized, and then, the spatial attributes (like source position, size and directivity) are applied by specific signal processing algorithms. In this paper we present an architecture that combines additive sound synthesis and 3D positional audio at the same level of sound generation. Our algorithm is based on inverse fast Fourier transform synthesis and amplitude-based sound positioning. It allows synthesizing and spatializing efficiently sinusoids and colored noise, to simulate point-like and extended sound sources. The audio rendering can be adapted to any reproduction system (headphones, stereo, 5.1 etc.). Possibilities offered by the algorithm are illustrated with environmental sounds.

7510
Harmonic Sinusoidal + Noise Modeling of Audio Based on Multiple F0 Estimation
Bartkowiak, Maciej; Zernicki, Tomasz
This paper deals with the detection and tracking of multiple harmonic series. We consider a bootstrap approach based on prior estimation of F0 candidates and subsequent iterative adjustment of a harmonic sieve with simultaneous refinement of the F0 and inharmonicity factor. Experiments show that this simple approach is an interesting alternative to popular strategies, where partials are detected without harmonic constraints, and harmonic series are resolved from mixed sets afterwards. The most important advantage is that common problems of tonal/noise energy confusion in case of unconstrained peak detection are avoided. Moreover, we employ a popular LP-based tracking method which is generalized to dealing with harmonically related groups of partials by using a vector inner product as the prediction error measure. Two alternative extensions of the harmonic model are also proposed in the paper that result in greater naturalness of the reconstructed audio: an individual frequency deviation component and a complex narrowband individual amplitude envelope.

7511
Sound Extraction of Delackered Records
Johnsen, Ottar; Bapst, Frédéric; Seydoux, Lionel
Most direct cut records are made of an aluminum or glass plate with a coated acetate lacquer. Such records are often crackled due to the shrinkage of the coating. It is impossible to read such records mechanically. We are presenting here a technique to reconstruct the sound from such record by scanning the image of the record and combining the sound from the different parts of the "puzzle". The system has been tested by extracting sounds from sound archives in Switzerland and in Austria. The concepts will be presented as well as the main challenges. Extracted sound samples will be played.

7512
Parametric Interpolation of Gaps in Audio Signals
Lukin, Alexey; Todd, Jeremy
The problem of interpolation of gaps in audio signals is important for the restoration of degraded recordings. Following the parametric approach over a sinusoidal model recently suggested in JAES by Lagrange et al., this paper proposes an extension to this interpolation algorithm by considering the interpolation of a noisy component in a "sinusoidal+noise" signal model. Additionally, a new interpolator for sinusoidal components is presented and evaluated. The new interpolation algorithm is suitable for a wider range of audio recordings than just the interpolation of a sinusoidal signal component.

7513
Classification of Musical Genres Using Audio Waveform Descriptors in MPEG-7
Osmanovic, Nermin
Automated genre classification makes it possible to determine the musical genre of an incoming audio waveform. One application of this is to help listeners find music they like more quickly among millions of tracks in an online music store. By using numerical thresholds and the MPEG-7 descriptors, a computer can analyze the audio stream for occurrences of specific sound events such as kick drum, snare hit, and guitar strum. The knowledge about sound events provides a basis for the implementation of a digital music genre classifier. The classifier inputs a new audio file, extracts salient features, and makes a decision about the musical genre based on the decision rule. The final classification results show a recognition rate in the range 75% - 94% for five genres of music

7514
Loudness Descriptors to Characterize Programs and Music Tracks
Skovenborg, Esben; Lund, Thomas
We present a set of key numbers to summarize loudness properties of an audio segment, broadcast program or music track: the loudness descriptors. The computation of these descriptors is based on a measurement of loudness level, such as specified by the ITU-R BS.1770. Two fundamental loudness descriptors are introduced: Center of Gravity and Consistency. These two descriptors were computed for a collection of audio segments from various sources, media and formats. This evaluation demonstrates that the descriptors can robustly characterize essential properties of the segments. We propose three different applications of the descriptors: for diagnosing potential loudness problems in ingest material; as a means for performing a quality check, after processing/editing; or for use in a delivery specification.

7515
Methods for Identification of the Tuning System in Audio Musical Signals
Heydarian, Peyman; Jones, Lewis; Seago, Allan
The tuning system is an important aspect of a piece. It specifies the scale intervals and is an indicator of the emotions of a musical file. There is a direct relationship between musical mode and the tuning of a piece for modal musical traditions. So, the tuning system carries valuable information, which is worth incorporating into metadata of a file. In this paper different algorithms for automatic identification of the tuning system are presented and compared. In the training process, spectral and chroma average, and pitch histograms, are used to construct reference patterns for each class. The same is done for the testing samples and a similarity measure like the Manhattan distance classifies a piece into different tuning classes.

7516
“Roughometer”: Realtime Roughness Calculation and Profiling
Villegas, Julian; Cohen, Michael
A software tool capable of determining auditory roughness in real-time is presented. This application, based on Pure-Data (Pd), calculates the roughness of audio streams using a spectral method originally proposed by Vassilakis. The processing speed is adequate for many realtime applications, and results indicate limited but significant agreement with an internet application of the chosen model. Finally, the usage of this tool is illustrated by the computation of a roughness profile of a musical composition that can be compared to its perceived patterns of `tension' and `relaxation.'

7517
Graceful Degradation for Digital Radio Mondiale (DRM)
Kraemer, Ferenc; Schuller, Gerald
A method is proposed that is able to maintain an adequate transmission quality of broadcasting programs over channels strongly impaired by fading. Although attempts of providing Graceful Degradation are manifold, the so called “brickwall effect” is inherent in most digital broadcasting system. The main concept of the proposed method focuses on the open standard Digital Radio Mondiale (DRM). Our approach is to introduce an additional low bit rate parallel backup audio stream alongside the main radio stream. This backup stream bridges occurring dropouts in the main stream. Two versions are evaluated. One uses the standardized HVXC speech codec for encoding the parallel backup audio stream. The other version additionally uses a specially developed sinusoidal music codec.

7518
Factors Affecting Perception of Audio-Video Synchronization in Television
Mason, Andrew; Salmon, Richard
The increasing complexity of television broadcasting, has, over the decades, resulted in an increased variety of ways in which audio and video can presented to the audience after experiencing different delays. This paper explores the factors that affect whether what is presented to the audience will appear to be correct. Experimental results of a study of the effect of video spatial resolution are included. Several international organizations are working to solve technical difficulties that result in incorrect synchronisation of audio and video. A summary of their activities is included. The Audio Engineering Society Standards Committee has a project to standardize an objective measurement method, and a test signal and prototype measurement apparatus contributed to the project are described.

7519
Absolute Threshold of Coherence of Position Perception between Auditory and Visual Sources for Dialogs
Munoz, Roberto; Recuero, Manuel; Duran, Diego; Gazzo, Manuel
Under certain conditions, auditory and visual information are integrated into a single unified perception, even when they originate from different locations in space. The main motivation for this study was to find the absolute perception threshold of position coherence between sound and image, when moving the image across the screen, and when panning the sound. In this manner, it is possible to subjectively quantify, by means of the constant stimulus psychophysical method [1], the maximum difference of position between sound and image considered coherent by a viewer of audiovisual productions. This paper discusses the accuracy necessary to match the position of the sound and its image on the screen. The results of this study could be used to develop sound mixing criteria for audiovisual productions.

7520
Clandestine Wireless Development During WWII
Paul, Jon
We describe the many advances in spy radios during and after WWII, starting with the huge suitcase B2 suitcase transceiver, through several stages of miniaturization and eventually down to small modules a few inches in size just after the War. A top secret set known as the S-Phone was a VHF full duplex radiotelephone. It was used by clandestine agents and Partisans for communications with ships and planes. The surprising sophistication and fast engineering development of these radios will be illustrated with photographs and schematics from the collection of the Crypto-Museum. This multimedia presentation has vintage 1940s music and original WWII propaganda posters.

7521
Application of Multichannel Impulse Response Measurement to Automotive Audio
de Vries, Diemer; Strauß, Michael
Audio reproduction in small enclosures holds a couple of differences in comparison to conventional room acoustics. Today’s car audio systems meet sophisticated expectations but still the automotive listening environment delivers critical acoustic properties. During the design of such an audio system it is helpful to gain insight into the temporal and spatial distribution of the acoustic field's properties. Because room acoustic modeling software reaches its limits the use of acoustic imaging methods can be seen as a promising approach. This paper describes the application of wave field analysis based on a multichannel impulse response measurement in an automotive use case. Besides a suitable preparation of the theoretical aspects, the analysis method is used to investigate the acoustic wave field inside a car cabin.

7522
Multichannel Low Frequency Room Simulation with Properly Modeled Source Terms – Multiple Equalization Comparison
Matheson, Ryan J.
At low frequencies unwanted room resonances in regular sized rectangular listening rooms cause problems. Various methods for reducing these resonances are available including some multi-channel methods. Thus with introduction of setups like 5.1 surround into home theatre systems there are now more options available to perform active resonance control using the existing speaker array. We focus primarily on comparing, separately, each step of speaker placement and its effects on the response in the room as well as the effect of adding additional symmetrically placed speakers in the rear to cancel out any additional room resonances. The comparison is done by use of a Finite Difference Time Domain (FDTD) simulator with focus on properly modeling a source in the simulation. A discussion about the ability of a standard 5.1 setup to utilize a multi-channel equalization technique (without adding additional speakers to the setup) and a modal equalization technique is later discussed.

7523
A Super-Wide-Range Microphone with Cardioid Directivity
Ono, Kazuho; Sugimoto, Takehiro; Ando, Akio; Nomura, Tomohiro; Chiba, Yutaka; Imanaga, Keishi
This paper describes a super-wide-range microphone with cardioid directivity, which covers the frequency range up to 100 kHz. The authors have successfully developed the omni-directional microphone capable of picking up sounds of up to 100 kHz with low noise. The proposed microphone uses an omni-directional capsule adopted in the omni-directional super-wide-range microphone and a bi-directional capsule which is newly designed to fit the characteristics of the omni-directional one. The output signals of both capsules are synthesized as the output signals to achieve cardioid directivity. The measurement results show that the proposed microphone achieves wide frequency range up to 100 kHz, as well as low noise characteristics and excellent cardioid directivity.

7524
Methods and Limitations of Line Source Simulation
Feistel, Stefan; Thompson, Ambrose; Ahnert, Wolfgang
Although line array systems are in widespread use today, investigations of the requirements and methods for accurate modeling of line sources are scarce. In previous publications the concept of the Generic Loudspeaker Library (GLL) was introduced. We show that on the basis of directional elementary sources with complex directivity data finite line sources can be simulated in a simple, general and precise manner. We derive measurement requirements and discuss the limitations of this model. Additionally, we present a second step of refinement, namely the use of different directivity data for cabinets of identical type based on their position in the array. All models are validated by measurements. We compare the approach presented with other proposed solutions.

7525
Can One Perform Quasi-Anechoic Measurements in Normal Rooms?
Vanderkooy, John; Lipshitz, Stanley
This paper is an analysis of two methods that attempt to achieve high resolution frequency responses at low frequencies from measurements made in normal rooms. Such data is contaminated by reflections before the low-frequency impulse response of the system has fully decayed. By modifying the responses to decay more rapidly, then windowing a reflection-free portion, and finally recovering the full response by deconvolution, these quasi-anechoic methods purport to thwart the usual reciprocal uncertainty relationship between measurement duration and frequency resolution. One method works by equalizing the response down to dc, the other by increasing the effective highpass corner frequency of the system. Each method is studied with simulations, and both appear to work to varying degrees, but we question whether they are measurements or effectively simply model extensions. In practice noise significantly degrades both procedures.

7526
Automatic Verification of Large Sound Reinforcement Systems Using Models of Loudspeaker Performance Data
Dalbjörn, Klas; Berg, Johan
A method is described to automatically verify individual loudspeaker integrity and confirm the proper configuration of amplifier-loudspeaker connections in sound reinforcement systems. Using impedance sensing technology in conjunction with software-based loudspeaker performance modeling, the procedure verifies that the load presented at each amplifier output corresponds to impedance characteristics as described in the DSP system’s currently loaded model. Accurate verification requires use of load impedance models created by iterative testing of numerous loudspeakers.

7527
Bend Radius
Lampen, Stephen; Dole, Carl; Wan, Shulamite
Designers, installers, and system integrators, have many rules and guidelines to follow. Most of these are intended to maximize cable and equipment performance. Many of these are ‘rules-of-thumb’, simple guidelines, easy to remember, and often just as easily broken. One of these is the ‘rule-of-thumb’ regarding the bending of cable, especially coaxial cable. Many may have heard the term “No tighter than ten times the diameter.” While this can be helpful, in a general way, there is a deeper and more complex question. What happens when you do bend cable? What if you have no choice? Often a specific choice of rack or configuration of equipment requires that cables be bent tighter than that recommendation. And what happens if you ‘unbend’ a cable that has been damaged? Does it stay damaged or can it be restored? This paper outlines a series of laboratory tests to determine exactly what happens when cable is bent and what the reaction is. Further, we will analyze the effect of bending on cable performance, specifically looking at impedance variations and return loss (signal reflection). For high-definition video signals (HD-SDI) return loss is the key to maximum cable length, bit errors, and open eye patterns. So analyzing the effecting of bending will allow us to determine signal quality based on the bending of an individual cable. But does this apply to digital audio cables? Does the relatively low frequencies of AES digital signals make a difference? Can these cables be bent with less effect on performance? These tests were repeated on both coaxial cable of different sizes, and twisted pairs. Flexible coax cables were tested, as well as the standard solid-core installation versions. Paired cables consisted of AES digital audio shielded cables, both install and flexible versions, were also tested.

7528
Detecting Changes in Audio Signals by Digital Differencing
Waslo, Bill
A software application has been developed to provide an accessible method, based on signal subtraction, to determine whether or not an audio signal may been perceptibly changed by passing through components, cables, or similar processes or treatments. The goals of the program, the capabilities required of it, its effectiveness and the algorithms it uses are described. The program is made freely available to any interested users for use in such tests.

7529
Research on a Measuring Method of Headphones and Earphones Using HATS
Inanaga, Kiyofumi; Hara, Takeshi; Rasmussen, Gunnar; Riko, Yasuhiro
Currently various types of couplers are used for measurement of headphones and earphones. The coupler was selected according to the device under test by the measurer. Accordingly it was difficult to compare the characteristics of headphones and earphones. A measuring method was proposed using HATS and a simulated program signal. However, the method had some problems in the shape of ear hole, and the measured results were not reproducible. We tried to improve the reproducibility of the measurement using several pinna models. As the result, we achieved a measuring platform using HATS, which gives good reproducibility of measured results for various types of headphones and earphones and then makes it possible to compare the measured results fairly.

7530
Loudspeaker Production Variance
Hutt, Steven; Fincham, Laurie
Numerous quality assurance philosophies have evolved over the last few decades designed to manage manufacturing quality. Managing quality control of production loudspeakers is particularly challenging. Variation of sub-components and assembly processes across loudspeaker driver production batches may lead to excessive variation of sensitivity, bandwidth, frequency response and distortion characteristics etc. As loudspeaker drivers are integrated into production audio systems these variants result in broad performance permutation from system to system that affects all aspects of acoustic balance and spatial attributes. This paper will discuss traditional electro-dynamic loudspeaker production variation.

7531
Distributed Mechanical Parameters Describing Vibration and Sound Radiation of Loudspeaker Drive Units
Klippel, Wolfgang; Schlechter, Joachim
The mechanical vibration of loudspeaker drive units is described by a set of linear transfer functions and geometrical data which are measured at selected points on the surface of the radiator (cone, dome, diaphragm, piston, panel) by using a scanning technique. These distributed parameters supplement the lumped parameters (T/S, nonlinear, thermal), simplify the communication between cone, driver and loudspeaker system design and open new ways for loudspeaker diagnostics. The distributed vibration can be summarized to a new quantity called accumulated acceleration level (AAL) which is comparable with the sound pressure level (SPL) if no acoustical cancellation occurs. This and other derived parameters are the basis for modal analysis and novel decomposition techniques which make the relationship between mechanical vibration and sound pressure output more transparent. Practical problems and indications for practical improvements are discussed for various example drivers. Finally, the usage of the distributed parameters within finite and boundary element analyses is addressed and conclusions for the loudspeaker design process are made.

7532
A New Methodology for the Acoustic Design of Compression Driver Phase Plugs with Radial Channels
Dodd, Mark; Oclee-Brown, Jack
Recent work by the authors [1] describes an improved methodology for the design of annular-channel, dome compression drivers. Although not so popular, radial channel phase plugs are used in some commercial designs. While there has been some limited investigation into the behaviour of this kind of compression driver [2], the literature is much more extensive for annular types. In particular, the modern approach to compression driver design, based on a modal description of the compression cavity, as first pioneered by Smith [3], has no equivalent for radial designs. In this paper we first consider if a similar approach is relevant to radial-channel phase plug designs. The acoustical behaviour of a radial-channel compression driver is analytically examined in order to derive a geometric condition that ensures minimal excitation of the compression cavity modes.

7533
Mechanical Properties of Ferrofluids in Loudspeakers
Lemarquand, Guy; Ravaud, Romain; Lemarquand, Valerie; Depollier, Claude
This paper describes the mechanical properties of ferrofluid seals in ironless electrodynamic loudspeakers. The motor is constituted of several outer stacked ring permanent magnets. The inner moving part is a piston. In addition, two ferrofluid seals are used which replace the classical suspension. Indeed, these seals fulfill several functions. First, they ensure the airtightness between the loudspeaker faces. Second, they act as bearings and center the moving part. Finally, the ferrofluid seals also exert a pull back force on the moving piston. Both radial and axial forces exerted on the piston are calculated thanks to analytical formulations. Furthermore, the shape of the seal is discussed as well as the optimal quantity of ferrofluid. The seal capacity is also calculated.

7534
An Ironless Low Frequency Subwoofer Functioning under Its Resonance Frequency
Merit, Benoit; Lemarquand, Guy; Nemoff, Bernard
A low frequency loudspeaker (10Hz-100Hz) is described. Its structure is totally ironless in order to avoid nonlinear effects due to the presence of iron. The large diaphragm and the high force factor of the loudspeaker lead to its high efficiency. Efforts have been made for reducing the non-linearities of the loudspeaker for a more accurate sound reproduction. In particular we have developed a motor totally made of permanent magnets, which create a uniform induction across the entire intended displacement of the coil. The motor linearity and the high force factor of this flat loudspeaker make it possible to function under its resonance frequency with a great accuracy.

7535
Line Arrays with Controllable Directional Characteristics — Theory and Practice
Fincham, Laurie; Brown, Peter
A so-called arc line array is capable of providing directivity control. Applying simple amplitude shading can, in theory, provide good off-axis lobe suppression and constant directivity over a frequency range determined at low-frequencies by line length and at high-frequencies by driver spacing. Array transducer design present additional challenges – the dual requirements of close spacing, for accurate high-frequency control, and a large effective radiating area, for good bass output, are incompatible with the use of multiple full-range drivers. A novel drive unit layout is proposed and theoretical and practical design criteria are presented for a two-way line with controllable directivity and virtual elimination of spatial aliasing. The PC-based array controller permits real-time changes in beam parameters for multiple overlaid beams.

7536
Loudspeaker Directivity Improvement Using Low Pass and All Pass Filters
Hughes, Charles
The response of loudspeaker systems employing multiple drivers within the same pass band is often less than ideal. This is due to the physical separation of the drivers and their lack of proper acoustical coupling within the higher frequency region of their use. The resultant comb filtering is sometimes addressed by applying a low pass filter to one or more of the drivers within the pass band. This can cause asymmetries in the directivity response of the loudspeaker system. A method is presented to greatly minimize these asymmetries through the use of low pass and all pass filters. This method is also applicable as a means to extend the directivity control of a loudspeaker system to lower frequencies.

7537
On the Necessary Delay for the Design of Causal and Stable Recursive Inverse Filters for Loudspeaker Equalization
Marques, Avelino; Freitas, Diamantino
The authors have developed and applied a novel approach to the equalization of non-minimum phase loudspeaker systems, based on the design of Infinite Impulse Response (recursive) inverse filters. In this paper the results and improvements attained on this novel IIR filter design method are presented. A special attention has been given to the delay of the equalized system. The boundaries to be posed on the search space of the delay for a causal and stable inverse filter, to be used in the nonlinear least squares minimization routine, are studied, identified and related with the phase response of a test system (to be equalized) and with the order of the inverse filter; finally, these observations and relations are extended and applied to multi-way loudspeaker systems, demonstrating the connection of the lower and upper bounds of the delay with the loudspeaker’s crossover filters phase response and inverse filter order.

7538
A Piano Sound Database for Testing Automatic Transcription Methods
Ortiz-Berenguer, Luis; Blanco-Martin, Elena; Alvarez-Fernandez, Alberto; Blas-Moncalvillo, Jose A.; Casajus-Quiros, Francisco J.
A piano sound database, called ‘PianoUPM’, is presented. It is intended to help the researching community in developing and testing transcription methods. A practical database needs to contain notes and chords played through the full piano range and it needs to be recorded from acoustic pianos rather than synthesized ones. The presented piano sound database includes the recording of 13 pianos from different manufacturers. There are both upright and grand pianos. The recordings include the eighty-eight notes and eight different chords played both in legato and staccato styles. It also includes some notes of every octave played with four different forces to analyze the nonlinear behavior. This work has been supported by the Spanish National Project TEC2006-13067-C03-01/TCM.

7539
Measurements of Spaciousness for Stereophonic Music
Sarroff, Andy; Bello, Juan P.
The spaciousness of pre-recorded stereophonic music, or how large and immersive the virtual space of it is perceived to be, is an important feature of a produced recording. Quantitative models of spaciousness as a function of a recording's (1) wideness of source panning and of a recording's (2) amount of overall reverberation are proposed. The models are independently evaluated in two controlled experiments. In one, the panning widths of a distribution of sources with varying degrees of panning are estimated; in the other, the extents of reverberation for controlled mixtures of sources with varying degrees of reverberation are estimated. The models are shown to be valid in a controlled experimental framework.

7540
Music Annotation and Retrieval System Using Anti-Models
Chen, Zhi-Sheng; Zen, Jia-Min; Jang, Jyh-Shing Roger
Query-by-semantic-description (QBSD) is a natural way for searching/annotating music in a large database. We propose such a system by considering anti-words for each annotation word based on the concept of supervised multi-class labeling (SML). Moreover, words that are highly correlated with the anti-semantic meaning of a word constitute its anti-word set. By modeling both a word and its anti-word set, our system can achieve +8.21% and +1.6% gains of average precision and recall against SML under the condition of an equal average number of annotation words, that is, 10. By incorporating anti-models, we also allow queries with anti-semantic words, which is not an option for previous systems.

7541
The Effects of Lossy Audio Encoding on Onset Detection Tasks
Jacobson, Kurt; Davies, Matthew; Sandler, Mark
In large audio collections, it is common to store audio content with perceptual encoding. However, encoding parameters may vary from collection to collection or even within a collection - using different bit rates, sample rates, codecs, etc. We evaluate the effect of various audio encodings on the onset detection task. We show that audio-based onset detection methods are surprisingly robust in the presence of MP3 encoded audio. Statistically significant changes in onset detection accuracy only occur at bit-rates lower than 32kbps.

7542
An Evaluation of Pre-Processing Algorithms for Rhythmic Pattern Analysis
Gruhne, Matthias; Dittmar, Christian; Gaertner, Daniel; Schuller, Gerald
For the semantic analysis of polyphonic music, such as genre recognition, rhythmic pattern features (also called Beat Histogram) can be used. Feature extraction is based on the correlation of rhythmic information from drum instruments in the audio signal. In addition to drum instruments, the sounds of pitched instruments are usually also part of the music signal to analyze. This can have a significant influence on the correlation patterns. This paper describes the influence of pitched instruments for the extraction of rhythmic features, and evaluates two different pre-processing methods. One method computes a sinusoidal and noise model, where its residual signal is used for feature extraction. In the second method, a drum transcription based on spectral characteristics of drum sounds is performed, and the rhythm pattern feature is derived directly from the occurrences of the drum events. Finally, the results are explained and compared in detail.

7543
A Framework for Producing Rich Musical Metadata in Creative Music Production
Fazekas, Gyorgy; Raimond, Yves; Sandler, Mark
Musical metadata may include references to individuals, equipment, procedures, parameters or audio features extracted from signals. There are countless possibilities for using this data during the production process. An intelligent audio editor, besides internally relying on it, can be both producer and consumer of information about specific aspects of music production. In this paper, we propose a framework for producing and managing meta information about a recording session, a single take or a subsection of a take. As basis for the necessary knowledge representation we use the Music Ontology with domain specific extensions. We provide examples on how metadata can be used creatively, and demonstrate the implementation of an extended metadata editor in a multitrack audio editor application.

7544
SoundTorch: Quick Browsing in Large Audio Collections
Heise, Sebastian; Hlatky, Michael; Loviscach, Jörn
Musicians, sound engineers, and foley artists face the challenge of finding appropriate sounds in vast collections containing thousands of audio files. Imprecise naming and tagging forces users to review dozens of files in order to pick the right sound. Acoustic matching is not necessarily helpful here as it needs a sound exemplar to match with and may miss relevant files. Hence, we propose to combine acoustic content analysis with accelerated auditioning: Audio files are automatically arranged in 2D by psychoacoustic similarity. A user can shine a virtual flashlight onto this representation; all sounds in the light cone are played back simultaneously, their position indicated through surround sound. User tests show that this method can leverage the human brain's capability to single out sounds from a spatial mixture and enhance browsing in large collections of audio content.

7545
File System Tricks for Audio Production
Hlatky, Michael; Heise, Sebastian; Loviscach, Jörn
Not every file presented by a computer operating system needs to be an actual stream of independent bits. We demonstrate that different types of virtual files and folders including so-called "Filesystems in Userspace" (FUSE) allow streamlining audio content management with relatively little additional complexity. For instance, an off-the-shelf database system may present a distributed sound library through (seemingly) standard files in a project-specific hierarchy with no physical copying of the data involved. Regions of audio files may be represented as separate files; audio effect plug-ins may be displayed as collections of folders for on-demand processing while files are read. We address differences between operating systems, available implementations, and lessons learned when applying such techniques.

7546
On the Minimum-Phase Nature of Head-Related Transfer Functions
Nam, Juhan; Kolar, Miriam A.; Abel, Jonathan S.
For binaural synthesis, head-related transfer functions (HRTFs) are commonly implemented as pure delays followed by minimum-phase systems. Here, the minimum-phase nature of HRTFs is studied. The cross-coherence between minimum-phase and unprocessed measured HRTFs was seen to be greater than 0.9 for a vast majority of the HRTFs, and was rarely below 0.8. Non-minimum-phase filter components resulting in reduced cross-coherence appeared in frontal and ipsilateral directions. The excess group delay indicates that these non-minimum-phase components are associated with regions of moderate HRTF energy. Other regions of excess phase correspond to high-frequency spectral nulls, and have little effect on cross-coherence.

7547
Apparatus Comparison for the Characterization of Spaces
Kestian, Adam; Roginska, Agnieszka
This work presents an extension of the Acoustic Pulse Reflectometry (APR) methodology that was previously used to obtain the characteristics of smaller acoustic spaces. Upon reconstructing larger spaces, the geometric configuration and characteristics of the measurement apparatus can be directly related to the clarity of the results. This paper describes and compares three measurement setups and apparatus configurations. The advantages and disadvantages of each methodology are discussed.

7548
Quantifying the Effect of Room Response on Automatic Speech Recognition Systems
Anderson, Jeremy; Harris, John
It has been demonstrated that the acoustic environment has an impact on timbre and speech intelligibility. Automatic speech recognition is an established area that suffers from the negative effects of mismatch between different room impulse responses (RIR.) To better understand the changes imparted by the RIR, we have created synthetic responses to simulate utterances recorded in different locations. Using speech recognition techniques to quantify our results, we then looked for trends in performance to connect with impulse response changes.

7549
In Situ Determination of Acoustic Absorption Coefficients
Mallais, Scott
The determination of absorption characteristics for a given material is developed for in situ measurements. Experiments utilize maximum length sequences and a single microphone. The sound pressure is modeled using the compact source approximation. Emphasis is placed on low frequency resolution which is dependent on both the geometry of the loudspeaker-microphone-sample configuration and the room in which the measurement is performed. Methods used to overcome this limitation are discussed. The concept of the acoustic center is applied in the low frequency region, modifying the calculation of the absorption coefficient.

7550
Head-Related Transfer Function Customization by Frequency Scaling and Rotation Shift Based on a New Morphological Matching Method
Guillon, Pierre; Guignard, Thomas; Nicol, Rozenn
Head-Related Transfer Functions (HRTFs) individualization is required to achieve high quality Virtual Auditory Spaces. An alternative to acoustic measurements is the customization of non-individual HRTFs. To transform HRTF data, we propose a combination of frequency scaling and rotation shift, whose parameters are predicted by a new morphological matching method. For six subjects, mesh models of head and pinnae are acquired, and differences in size and orientation of the pinnae are evaluated with a modified Iterative Closest Point (ICP) algorithm. Optimal HRTF transformations are computed in parallel. A relatively good correlation between morphological and transformation parameters is found and allows to predict the customization parameters from the registration of pinna shapes. The resulting model achieves better customization than frequency scaling only, which shows that adding the rotation degree of freedom improves HRTF individualization.

7551
An Investigation of 2-D Multizone Surround Sound Systems
Poletti, Mark
Surround sound systems can produce a desired sound field over an extended region of space by using higher order Ambisonics. One application of this capability is the production of multiple independent soundfields in separate zones. This paper investigates multi-zone surround systems for the case of two dimensional reproduction. A least squares approach is used for deriving the loudspeaker weights for producing a desired single frequency wave field in one of N zones, while producing silence in the other N-1 zones. It is shown that reproduction in the active zone is more difficult when an inactive zone is in-line with the virtual sound source and the active zone. Methods for controlling this problem are discussed.

7552
Two-Channel Matrix Surround Encoding for Flexible Interactive 3-D Audio Reproduction
Jot, Jean-Marc
The two-channel matrix surround format is widely used for connecting the audio output of a video gaming system to a home theater receiver for multichannel surround reproduction. This paper describes the principles of a computationally-efficient interactive audio spatialization engine for this application. Positional cues including 3-D elevation are encoded for each individual sound source by frequency-independent inter-channel phase and amplitude differences, rather than HRTF cues. A matrix surround decoder based on frequency-domain Spatial Audio Scene Coding (SASC) is able to faithfully reproduce both ambient reverberation and positional cues over headphones or arbitrary multi-channel loudspeaker reproduction formats, while preserving source separation despite the intermediate encoding over only two channels.

7553
Is My Decoder Ambisonic?
Heller, Aaron; Lee, Richard; Benjamin, Eric
In earlier papers, the present authors established the importance of various aspects of Ambisonic decoder design: a decoding matrix matched to the geometry of the loudspeaker array in use, phase-matched shelf filters, and near field compensation. These are needed for accurate reproduction of spatial localization cues, such as interaural time difference (ITD), interaural level difference (ILD), and distance cues. Unfortunately, many listening tests of Ambisonic reproduction reported in the literature either omit the details of the decoding used or utilize suboptimal decoding. In this paper we review the acoustic and psychoacoustic criteria for Ambisonic reproduction; present a methodology and tools for ``black box' testing to verify the performance of a candidate decoder; and present and discuss the results of this testing on some widely used decoders.

7554
Exploiting Human Spatial Resolution in Surround Sound Decoder Design
Moore, David; Wakefield, Jonathan
This paper presents a technique whereby the localisation performance of surround sound decoders can be improved in directions in which human hearing is more sensitive to sound source location. Research into the Minimum Audible Angle is explored and incorporated into a fitness function based upon a psychoacoustic model. This fitness function is used to guide a heuristic search algorithm to design new Ambisonic decoders for a 5-speaker surround sound layout. The derived decoder is successful in matching the variation in localisation performance of the human listener with better performance to the front and rear and reduced performance to the sides. The effectiveness of the standard ITU 5-speaker layout versus a non-standard layout is also considered in this context.

7555
Surround System Based on Three-Dimensional Sound Field Reconstruction
Fazi, Filippo; Nelson, Philip; Christensen, Jens E.; Seo, Jeongil
The theoretical fundamentals and the simulated and experimental performance of an innovative surround sound system are presented. The proposed technology is based on the physical reconstruction of a three dimensional target sound field over a region of the space using an array of loudspeakers surrounding the listening area. The computation of the loudspeaker gains includes the numerical or analytical solution of an integral equation of the first kind. The experimental setup and the measured reconstruction performance of a system prototype constituted by a three dimensional array of 40 loudspeakers are described and discussed.

7556
A Comparison of Wave Field Synthesis and Higher-Order Ambisonics with Respect to Physical Properties and Spatial Sampling
Spors, Sascha; Ahrens, Jens
Wave field synthesis (WFS) and higher-order Ambisonics (HOA) are two high-resolution spatial sound reproduction techniques aiming at overcoming some of the limitations of stereophonic reproduction techniques. In the past, the theoretical foundations of WFS and HOA have been formulated in a quite different fashion. Although, some work has been published that aims at comparing both approaches their similarities and differences are not well documented. This paper formulates the theory of both approaches in a common framework, highlights the different assumptions made to derive the driving functions and the resulting physical properties of the reproduced wave field. Special attention will be drawn to the consequences of spatial sampling since both approaches differ significantly here.

7557
Reproduction of Virtual Sound Sources Moving at Supersonic Speeds in Wave Field Synthesis
Ahrens, Jens; Spors, Sascha
In conventional implementations of wave field synthesis, moving sources are reproduced as sequences of stationary positions. As reported in the literature, this process introduces various artifacts. It has been shown recently that these artifacts can be reduced when the physical properties of the wave field of moving virtual sources are explicitly considered. However, the findings were only applied to virtual sources moving at subsonic speeds. In this paper we extend the published approach to the reproduction of virtual sound sources moving at supersonics speeds. The properties of the actually reproduced sound field are investigated via numerical simulations.

7558
An Efficient Method to Generate Particle Sounds in Wave Field Synthesis
Beckinger, Michael; Brix, Sandra
Rendering a couple of virtual sound sources for wave field synthesis (WFS) in real time is nowadays feasible using the calculation power of state-of-the-art personal computers. If immersive atmospheres containing thousands of sound particles like rain and applause should be rendered in real time for a large listening area with a high spatial accuracy, calculation complexity increases enormously. A new algorithm based on continuously generated impulse responses and following convolutions, which renders many sound particles in an efficient way will be presented in this paper. The algorithm was verified by first listening tests and its calculation complexity was evaluated as well.

7559
Audibility of Phase Response Differences in a Stereo Playback System. Part 2: Narrow-Band Stimuli in Headphones and Loudspeakers
Choisel, Sylvain; Martin, Geoff
An series of experiments were conducted in order to measure the audibility thresholds of phase differences between channels using mismatched cross-over networks. In Part 1 of this study, it was shown that listeners are able to detect very small inter-channel phase differences when presented with wide-band stimuli over headphones, and that the threshold was frequency dependent. This second part of the investigation focuses on listeners' abilities with narrow-band signals (from 63 to 8000\,Hz) in headphones as well as loudspeakers. The results confirm the frequency dependency of the audibility threshold over headphones, whereas for loudspeaker playback the threshold was essentially independent of the frequency.

7560
Time Variance of the Suspension Nonlinearity
Agerkvist, Finn; Pedersen, Bo Rhode
It is well known that the resonance frequency of a loudspeaker depends on how it is driven before and during the measurement. Measurement done right after exposing it to high levels of electrical power and/or excursion giver lower values than what can be measured when the speaker is ‘cold’. This paper investigates the changes in compliance the driving signal can cause, this includes low level short duration measurements of the resonance frequency as well as high power long duration measurements of the non-linearity of the suspension. It is found that at low levels the suspension softens but recovers quickly. The the high power and long term measurements affect the non-linearity of the speaker, by incresing the compliance value for all values of displacement. This level dependency is validated with distortion measurements and it is demonstrated how improved accuracy of the non-linear model can be obtained by including the level dependency.

7561
A Study of the Creep Effect in Loudspeaker Suspension
Agerkvist, Finn; Thorborg, Knud; Tinggard, Carsten
This paper investigates the creep effect, the visco elastic behaviour of loudspeaker suspension parts, which can be observed as an increase in displacement far below the resonance frequency. The creep effect means that the suspension cannot be modelled as a simple spring. The need for an accurate creep model is even larger as the validity of loudspeaker models are now sought extended far into the nonlinear domain of the loudspeaker. Different creep models are investigated and implemented both in simple lumped parameter models as well as time domain non-linear models, the simulation results are compared with a series of measurements on three version of the same loudspeaker with different thickness and rubber type used in the surround.

7562
The Influence of Acoustic Environment on the Threshold of Audibility of Loudspeaker Resonances
Uprichard, Shelley; Choisel, Sylvain
Resonances in loudspeakers can produce a detrimental effect on sound quality. The reduction or removal of unwanted resonances has therefore become a recognized practice in loudspeaker tuning. This paper presents the results of a listening test which has been used to determine the audibility threshold of a single resonance in different acoustic environments: headphones, loudspeakers in a standard listening room and loudspeakers in a car. Real loudspeakers were measured and the resonances modelled as IIR filters. Results show that there is a significant interaction between acoustic environment and programme material.

7563
Confirmation of Chaos in a Loudspeaker System Using Time Series Analysis
Reiss, Joshua; Djurek, Ivan; Petosic, Antonio; Djurek, Danijel
The dynamics of an experimental electrodynamic loudspeaker is studied by using the tools of chaos theory and time series analysis. Delay time, embedding dimension, fractal dimension and other empirical quantities are determined from experimental data. Particular attention is paid to issues of stationarity in the system in order to identify sources of uncertainty. Lyapunov exponents and fractal dimension are measured using several independent techniques. Results are compared in order to establish independent confirmation of low dimensional dynamics and a positive dominant Lyapunov exponent. We thus show that the loudspeaker may function as a chaotic system suitable for low dimensional modeling and the application of chaos control techniques.

7564
Testing Loudness Models—Real vs. Artificial Content
Johnston, James
A variety of loudness models have been recently proposed and tested by various means. In this paper, some basic properties of loudness are examined, and a set of artificial signals are designed to test the "loudness space" based on principles dating back to Harvey Fletcher, or arguably to Wegel and Lane. Some of these signals, designed to model "typical" content, seem to reinforce the results of prior loudness model testing. Other signals, less typical of standard content, seem to show that there are some substantial differences when these less common signals and signal spectra are used.

7565
Audibility of High Q-Factor All-Pass Components in Head-Related Transfer Functions
Toledo, Daniela; Møller, Henrik
Head-related transfer functions (HRTFs) can be decomposed into minimum phase, linear phase and all-pass components. It is known that low Q-factor all-pass sections in HRTFs are audible as lateral shifts when the interaural group delay at low frequencies is above 30 microseconds. The goal of our investigation is to test the audibility of high Q-factor all-pass components in HRTFs and the perceptual consequences of removing them. A three-alternative forced choice experiment has been conducted. Results suggest that high Q-factor all-pass sections are audible when presented alone, but inaudible when presented with their minimum phase HRTF counterpart. It is concluded that high Q-factor all-pass sections can be discarded in HRTFs used for binaural synthesis.

7566
A Psychoacoustic Measurement and ABR for the Sound Signals in the Frequency Range Between 10 kHz and 24 kHz
Omata, Mizuki; Ashihara, Kaoru; Kyouso, Masaki; Koubori, Motoki; Moriya, Yoshitaka; Kyouso, Masaki; Kiryu, Shogo
In high definition audio media such as SACD and DVD-audio, wide frequency range far beyond 20 kHz is used. However, the auditory characteristics for the frequencies higher than 20 kHz have not been necessarily understood. At the first step to make clear the characteristics, we conducted a psycho-acoustic and a auditory brain-stem response(ABR) measurement for the sound signals in the frequency range between 10 kHz and 24 kHz. At the frequency of 22 kHz, the minimum hearing threshold in the psycho-acoustic measurement was 80 dB. At the frequency of 22 kHz, the thresholds of 100 dB in the ABR measurement could be measured for 1 of the 5 subjects.

7567
Quantifying the Strategy Taken by a Pair of Ensemble Hand-Clappers under the Influence of Delay
Darabi, Nima; Svensson, Peter; Farner, Snorre
Pairs of subjects were placed in two acoustically isolated rooms clapping together under an influence of delay up to 68 ms. Their trials were recorded and analyzed based on a definition of compensation factor or CF. This parameter was calculated from the recorded observations for both performers as a discrete function of time and thought of as a measure of the strategy taken by the subjects while clapping. Increasing the delay CF was shown to be increased linearly as it is desired to avoid tempo decrease for such high latencies. Theoretically a critical value for CF was defined as tempo over measure (or beat) duration and was used to explain why very short latencies may lead to a tempo acceleration in accordance with Chafe effect.

7568
Quantitative and Qualitative Evaluations for TV Advertisements Relative to the Adjacent Programs
Miyasaka, Eiichi; Kimura, Akiko
The sound levels of advertisements (CMs) in Japanese conventional terrestrial analogue broadcasting (TAB) were quantitatively compared with those in Japanese terrestrial digital broadcasting (TDB). The results show that the averaged CM-sound level in TDB was about 2 dB lower and the averaged standard deviation was wider than those in TAB, while there were few differences between TAB and TDB at some TV station. Some CMs in TDB were perceived clearly louder than the adjacent programs although the sound level differences between the CMs and the programs were only within ?}2 dB. Next, insertion methods of CMs into the main programs in Japan were qualitatively investigated. The results show that some kinds of the methods could unacceptably irritate viewers.

7569
Imperfections and Possible Advances in Analog Summing Amplifier Design
Kovinic, Milan; Drincic, Dragan; Jankovic, Sasha
The major requirement in the design of the analogue summing amplifier is the quality of the summing bus. The key problem in most common designs is the artefact of summing bus impedance, which cannot be considered as true physical impedance, because it has been generated by negative feedback. The loop gain of the amplifier used will limit the performance at higher audio frequencies where the loop gain is lower, increasing the channels cross talk. The inevitable effect of heavy feedback is the increased susceptibility of the amplifier to oscillate as well as sensitivity to RFI. The advanced solution, presented in this work, could be seen in the usage of the transistor common-base pair (CB-CB) configuration as a summing bus. The CB pair offers inherent low-input impedance, low-noise, very good frequency response, and, very importantly, makes the application of total feedback not necessarily.

7570
A Switchmode Power Supply Suitable for Audio Power Amplifiers
Gordon, Jay
Power supplies for audio amplifiers have different requirements than typical commercial power supplies. A tabulation of power supply parameters that affect the audio application is presented and discussed. Different types ofaudio amplifiers are categorized and shown to have different requirements. Over time new technologies have emerged which affect the implementation of AC to DC converters used in audio amplifiers. A brief history of audio power supply technology is presented. The evolution of a newly proposed interleaved boost with LLC resonant half bridge topology from preceding technologies is shown. The operation of the new topology is explained and its advantages are shown by a simulation of the circuit.

7571
On the Optimization of Enhanced Cascode
Danyuk, Dimitri
Twenty years ago enhanced cascode and other circuit topologies based on the same design principles were presented to audio amplifier designers. The circuit was supposed to be incorporated in transconductance gain stages and current sources. Enhanced cascode was used in some commercial products but have not received wide adoption. It was speculated that enhanced cascode has reduced phase margin and at times higher distortion being compared to conventional cascode. Enhanced cascode is analyzed on the basis of distortion and frequency response. It is shown how to make the most of enhanced cascode. Optimized novel circuit topology is presented.

7572
An Active Load and Test Method for Evaluating the Efficiency of Audio Power Amplifiers
Dymond, Harry; Mellor, Phil
This paper presents the design, implementation and use of an “active load” for audio power amplifier efficiency testing. The active load can simulate linear complex loads representative of real-world amplifier operation with a load modulus between 4 and 50 ohms inclusive, load phase-angles between −60° and +60° inclusive, and operates from 20 to 20,000 Hz. The active load allows for the development of an automated test procedure for evaluating the efficiency of an audio power amplifier across a range of output voltage amplitudes, load configurations and output signal frequencies. The results of testing a class-B and a class-D amplifier, each rated at 100 watts into 8 ohms are presented.

7573
An Objective Method of Measuring Subjective Click-and-Pop Performance for Audio Amplifiers
Christman (Schmidt), Kymberly
Click-and-Pop refers to any ‘clicks’ and ‘pops’ or other unwanted, audio-band transient signals that are reproduced by headphones or speakers when the audio source is turned on or off. Until recently, the industry’s characterization of this undesirable effect has been almost purely subjective. Marketing phrases such as ‘low pop noise’ and ‘clickless/popless operation’ illustrate the subjectivity applied in quantifying click-and-pop performance. The following paper presents a method that objectively quantifies this parameter, allowing meaningful, repeatable comparisons to be drawn between different components. Further, results of a subjective click-and-pop listening test are presented to provide a baseline for objectionable click-and-pop levels in headphone amplifiers.

7574
Effective Car Audio System Enabling Individual Signal Processing Operations of Coincident Multiple Audio Sources through Single Digital Audio Interface Line
Yoo, Chul-Jae; Ryu, In-Sik
There are three major audio sources in recent car environments: Primary audio (usually music including radio), Navigation voice prompt and Hands-free voice. Listening situations in cars include not only listening single audio source, but also listening concurrent multiple audio sources – for example, navigation guided as listening music and navigation guided or listening music as talking on a hands-free cell phone. In this paper, firstly, a conventional external amplifier system connected with a head unit by three audio interface lines was introduced. Then, effective automotive audio system having single SPDIF interface line which is capable of concurrent processing of above three kinds of audio sources was proposed. New system leads to reduce wire harness in car environments and also increases voice qualities by transmitting voice signals via an SPDIF digital line compared with that via analog lines.

7575
Digital Equalization of Automotive Sound Systems Employing Spectral Smoothed FIR Filters
Binelli, Marco; Farina, Angelo
In this paper we investigate about the usage of spectral smoothed FIR filters for equalizing a car audio system. The target is also to build short filters that can be processed on DSP processors with limited computing power. The inversion algorithm is based on the Nelson-Kirkeby method and on independent phase and magnitude smoothing, by means of a continuous phase method as Panzer and Ferekidis showed. The filter is aimed to create a "target" frequency response, not necessarily flat, employing a little number of taps and maintaining good performances everywhere inside the car's cockpit. As shown also by listening tests, smoothness and the choice of the right frequency response increase the performances of the car audio systems

7576
Implementation of a Generic Algorithm on Various Automotive Platforms
Esnault, Thomas; Raczinski, Jean-Michel
This paper describes a methodology to adapt a generic automotive algorithm to various embedded platforms while keeping the same audio rendering. To get over the limitations of the target DSPs, we have developed tools to control the transition from one platform to another including algorithm adaptation and coefficient computing. Objective and subjective validation processes allow us to certify the quality of the adaptation. With this methodology, productivity has been increased in an industrial context.

7577
Advanced Audio Algorithms for a Real Automotive Digital Audio System
Cecchi, Stefania; Palestini, Lorenzo; Peretti, Paolo; Moretti, Emanuele; Piazza, Francesco; Lattanzi, Ariano; Bettarelli, Ferruccio
In this paper an innovative modular digital audio system for car entertainment is proposed. The system is based on a plugin-based software (real-time) framework allowing reconfigurability and flexibility. Each plugin is dedicated to a particular audio task such as equalization and crossover filtering, implementing innovative algorithms. The system has been tested on a real car environment, with a hardware platform comprising professional audio equipments, running on a PC. Informal listening tests have been performed to validate the overall audio quality, and satisfactory results were obtained.

7578
Individual Subjective Preferences for the Relationship between SPL and Different Cinema Shot Sizes
Munoz, Roberto; Recuero, Manuel; Gazzo, Manuel; Duran, Diego
The main motivation for this study was to find Individual Subjective Preferences (ISP) for the relationship between SPL and different cinema shot sizes. By means of the psychophysical method of Adjustment (MA) [1], the preferred SPL for four of the most frequently used shot sizes, i.e., long shot, medium shot, medium close-up, and close-up, was subjectively quantified. Also using the Constant Stimulus Method [1], the preferred difference of SPL for different combinations of the above-mentioned shot sizes was studied. The results of this study could be used to develop sound mixing criteria for audiovisual productions.

7579
Improvements to a Spherical Binaural Capture Model for Objective Measurement of Spatial Impression with Consideration of Head Movements
Kim, Chungeun; Mason, Russell; Brookes, Tim
This research aims, ultimately, to develop a system for the objective evaluation of spatial impression, incorporating the finding from a previous study that head movements are naturally made in its subjective evaluation. A spherical binaural capture model, comprising a head-sized sphere with multiple attached microphones, has been proposed. Research already conducted found significant differences in interaural time and level differences, and cross-correlation coefficient, between this spherical model and a head and torso simulator. It is attempted to lessen these differences by adding to the sphere a torso and simplified pinnae. Further analysis of the head movements made by listeners in a range of listening situations determines the range of head positions that needs to be taken into account. Analyses of these results inform the optimum positioning of the microphones around the sphere model.

7580
Predicting Perceived Off-Center Sound Degradation in Surround Loudspeaker Setups for Various Multichannel Microphone Techniques
Peters, Nils; Giordano, Bruno L.; Kim, Sungyoung; Braasch, Jonas; McAdams, Stephen
Multiple listening tests were conducted to examine the influence of microphone techniques on the quality of sound reproduction. Generally, testing focuses on the central listening position (CLP), and neglects off-center listening positions. Exploratory tests focusing on the degradation in sound quality at off-center listening positions were presented at the 123rd AES Convention. Results showed that the recording technique does influence the degree of sound degradation at off-center positions. This paper focuses on the analysis of the binaural re-recording at the different listening positions in order to interpret the results of the previous listening tests. Multiple linear regression is used to create a predictive model which accounts for 85% of the variance in the behavioral data. The primary successful predictors were spectral and the secondary predictors were spatial in nature.

7581
Rapid Learning of Subjective Preference in Equalization
Sabin, Andrew; Pardo, Bryan
We describe and test an algorithm to rapidly learn a listener’s desired equalization curve. First, a sound is modified by a series of equalization curves. After each modification, the listener indicates how well the current sound exemplifies a target sound descriptor (e.g., “warm”). After rating, a weighting function is computed where the weight of each channel (frequency band) is proportional to the slope of the regression line between listener responses and within-channel gain. Listeners report that sounds generated using this function capture their intended meaning of the descriptor. Machine ratings generated by computing the similarity of a given curve to the weighting function are highly correlated to listener responses, and asymptotic performance is reached after only ~25 listener ratings.

7582
An Initial Validation of Individualized Crosstalk Cancellation Filters for Binaural Perceptual Experiments
Moore, Alastair; Tew, Anthony; Nicol, Rozenn
Crosstalk cancellation provides a means of delivering binaural stimuli to a listener for psychoacoustic research which avoids many of the problems of using headphone in experiments. The aim of this study was to determine whether individual crosstalk cancellation filters can be used to present binaural stimuli which are perceptually indistinguishable from a real sound source. The fast deconvolution with frequency dependent regularisation method was used to design crosstalk cancellation filters. The reproduction loudspeakers were positioned at ±90 degrees azimuth and the synthesised location was 0 degrees azimuth. Eight listeners were tested with three types of stimuli. In twenty-two out of the twenty-four listener/stimulus combinations there were no perceptible differences between the real and virtual sources. The results suggest that this method of producing individualised crosstalk cancellation filters is suitable for binaural perceptual experiments.

7583
Reverberation Echo Density Psychoacoustics
Huang, Patty; Abel, Jonathan S.; Terasawa, Hiroko; Berger, Jonathan
A series of psychoacoustic experiments were carried out to explore the relationship between an objective measure of reverberation echo density, called the normalized echo density (NED), and subjective perception of the time-domain texture of reverberation. In one experiment, 25 subjects evaluated the dissimilarity of signals having static echo densities. The reported dissimilarities matched absolute NED differences with an R2 of 93%. In a 19-subject experiment, reverberation impulse responses having evolving echo densities were used. With an R2 of 90%, the absolute log ratio of the late field onset times matched reported dissimilarities between impulse responses. In a third experiment, subjects consistently reported breakpoints in echo pattern character at NEDs at 0.3 and 0.7.

7584
Optimal Modal Spacing and Density for Critical Listening
Fazenda, Bruno; Wankling, Matthew
This paper presents a study on the subjective effects of modal spacing and density. These are measures often used as indicators to define particular aspect ratios and source positions to avoid low frequency reproduction problems in rooms. These indicators imply a given modal spacing leading to a supposedly less problematic response for the listener. An investigation into this topic shows that subjects can identify an optimal spacing between two resonances associated with a reduction of the overall decay. Further work to define a subjective counterpart to the Schroeder Frequency has revealed that an increase in density may not always lead to an improvement, as interaction between mode-shapes results in serious degradation of the stimulus, which is detectable by listeners.

7585
The Illusion of Continuity Revisited on Filling Gaps in the Saxophone Sound
Kleczkowski, Piotr
Some time-frequency gaps were cut from a recording of a motif played legato on the saxophone. Subsequently, the gaps were filled with various sonic material: noises and sounds of an accompanying band. The quality of the saxophone sound processed in this way was investigated by listening tests. In all of the tests, the saxophone seemed to continue through the gaps, an impairment in quality being observed as a change in the tone colour or an attenuation of the sound level. There were two aims of this research. First, to investigate whether the continuity illusion contributed to this effect, and second, to discover what kind of sonic material filling the gaps would cause the least deterioration in sound quality.

7586
The Incongruency Advantage for Sounds in Natural Scenes
Gygi, Brian; Shafiro, Valeriy
This research tests identification of environmental sounds (dogs barking or cars honking) in familiar auditory background scenes (street ambience, restaurants). Initial results with subjects trained on both the background scenes and the sounds to be identified showed a significant advantage of about 5% better identification accuracy for sounds that were incongruous with the background scene (e.g., a rooster crowing in a hospital). Studies with naïve listeners showed this effect is level-dependent: there is no advantage for incongruent sounds up to a Sound/Scene ratio (So/Sc) of -7.5 dB, after which there is again about 5% better identification. Modeling using spectral-temporal measures showed that saliency based on acoustic features cannot account for this difference.

7588
Advanced Passive Loudspeaker Protection
Dorsey, Scott
In a follow-on to a previous conference paper the author explores the use of polymeric positive temperature coefficient(PPTC) protection devices which have an discontinuous I/V curve that is theresult of a physical state change. He gives a simple model for designing networks employing incandescent lamps and PPTC devices together to give linear operation at low levels while providing effective limiting athigher levels to prevent speaker damage. Some discussion of applicationsin current service is provided.

7589
Target Modes in Moving Assemblies of Compression Drivers and Other Speakers
Bolaños, Fernando; Seoane, Pablo
The paper deals how to find the important modes in a moving assembly of compression drivers and other loudspeakers. Dynamic importance is an essential tool for those who work on modal analysis of systems with many degrees of freedom and complex structures. The important modes calculation or measurement in moving assemblies is an objective (absolute) method to find the relevant modes which acts on the dynamics of these transducers. Paper deals about axial modes and breath modes which are basic for loudspeakers. The model generalized masses and the participation factors are useful tools to find the moving assemblies important modes (target modes). The strain energy of the moving assembly, which represents the amount of available potential energy, is essential as well.

7590
Determining Manufacture Variation in Loudspeakers Through Measurement of Thiele/Small Parameters
Laurin, Scott; Reichard, Karl
Thiele/Small parameters have become a standard for characterizing loudspeakers. Using fairly straightforward methods, the Thiele/Small parameters for twenty nominally identical loudspeakers were determined. The data were compiled to determine the manufacturing variations. Manufacturing tolerances can have a large impact on the variability and quality of loudspeakers produced. Generally, when more stringent tolerances are applied, there is less variation and drivers become more expensive. Now that the loudspeakers have been characterized, each one will be driven to failure. Some loudspeakers will be intentionally degraded to accelerate failures. The goal is to correlate variation in the Thiele/Small parameters with variation in speaker failure modes and operating life.

7591
About Phase Optimization in Multitone Excitations
Bard, Delphine; Meyer, Vincent
Multitone signals are often used as excitation for the characterization of audio systems. The frequency spectrum of the response consists of harmonics of the frequencies contained in the excitation and intermodulation products. Beside the choice of frequencies, in order to avoid frequency overlapping, there is also the need to choose adequate magnitudes and phases for the different components that constitute the multitone signal. In this study, we will investigate how the choice of the phases will impact the properties of the multitone signal, but also how it will affect the performances of a compensation method based on Volterra kernels and using multitone signals as an excitation.

7592
Viscous Friction and Temperature Stability of the Mid-High Frequency Loudspeaker
Djurek, Ivan; Petosic, Antonio; Djurek, Danijel
Mid-high frequency loudspeakers behave quite differently as compared to low-frequency units, regarding effects coming from the surrounding air medium. Previous work stressed high influence of the imaginary part of the viscous force, which significantly affects the resonance frequency of mid-high frequency loudspeakers. Viscous force is relatively high dependent on temperature and humidity of the surrounding air, and in this paper we have evaluated how changes in temperature and humidity reflect to the loudspeaker's linearity, which may be significant for the quality of sound reproduction.

7593
Calorimetric Evaluation of Intrinsic Friction in the Loudspeaker Membrane
Petosic, Antonio; Djurek, Ivan; Djurek, Danijel
Friction losses in the vibrating system of an electrodynamic loudspeaker are represented by the intrinsic friction Ri which enters the equation of motion, and these losses are accompanied by irreversible release of the heat. A method is proposed for measurement of the friction losses in the loudspeaker's membrane by measurement of the thermocouple temperature probe glued to the membrane. Temperature on the membrane surface fluctuates stochastically as a result of thermo-elastic coupling in the membrane material. Evaluation of the amplitude in the temperature fluctuations enables an absolute and direct evaluation of intrinsic friction Ri entering friction force F=Ri•v(x), irrespective of the nonlinearity type and strength associated with the loudspeaker operation.

7594
Phantom Powering the Modern Condenser Microphone: A Practical Look at Conditions for Optimized Performance
Zaim, Mark; Kikutani, Tadashi; Green, Jackie
Phantom Powering a microphone is a decades old concept with powering conventions and methods that may have become obsolete, ineffective, or inefficient. Modern sound techniques, including those of live sound settings, now use many condenser microphones in settings that were previously dominated by dynamics. As a prerequisite for considering a modern phantom power specification or method, we study the efficiencies and requirements of microphones in typical multiple mic and high SPL settings in order to gain understanding of circuit and design requirements for the maximum dynamic range performance.

7595
QESTRAL (Part 1): Quality Evaluation of Spatial Transmission and Reproduction Using an Artificial Listener
Rumsey, Francis; Zielinski, Slawomir; Jackson, Philip; Dewhirst, Martin; Conetta, Robert; George, Sunish; Bech, Søren; Meares, David
Most current perceptual models for audio quality have so far tended to concentrate on the audibility of distortions and noises that mainly affect the timbre of reproduced sound. The QESTRAL model, however, is specifically designed to take account of distortions in the spatial domain such as changes in source location, width and envelopment. It is not aimed only at codec quality evaluation but at a wider range of spatial distortions that can arise in audio processing and reproduction systems. The model has been calibrated against a large database of listening tests designed to evaluate typical audio processes, comparing spatially degraded multichannel audio material against a reference. Using a range of relevant metrics and a sophisticated multivariate regression model, results are obtained that closely match those obtained in listening tests.

7596
QESTRAL (Part 2): Calibrating the QESTRAL Model Using Listening Test Data
Conetta, Robert; Rumsey, Francis; Zielinski, Slawomir; Jackson, Philip; Dewhirst, Martin; Bech, Søren; Meares, David; George, Sunish
The QESTRAL model is a perceptual model that aims to predict changes to spatial quality of service between a reference system and an impaired version of the reference system. To achieve this, the model required calibration using perceptual data from human listeners. This paper describes the development, implementation and outcomes of a series of listening experiments designed to investigate the spatial quality impairment of 40 processes. Assessments were made using a multi-stimulus test paradigm with a label-free scale, where only the scale polarity is indicated. The tests were performed at two listening positions, using experienced listeners. Results from these calibration experiments are presented. A preliminary study on the process of selecting of stimuli is also discussed.

7597
QESTRAL (Part 3): System and Metrics for Spatial Quality Prediction
Jackson, Philip; Dewhirst, Martin; Conetta, Robert; Zielinski, Slawomir; Rumsey, Francis; Meares, David; Bech, Søren; George, Sunish
The QESTRAL project aims to develop an artificial listener for comparing the perceived quality of a spatial audio reproduction against a reference reproduction. This paper presents implementation details for simulating the acoustics of the listening environment and the listener's auditory processing. Acoustical modelling is used to calculate binaural signals and simulated microphone signals at the listening position, from which a number of metrics corresponding to different perceived spatial aspects of the reproduced sound field are calculated. These metrics are designed to describe attributes associated with location, width and envelopment attributes of a spatial sound scene. Each provides a measure of the perceived spatial quality of the impaired reproduction compared to the reference reproduction. As validation, individual metrics from listening test signals are shown to match closely subjective results obtained, and can be used to predict spatial quality for arbitrary signals.

7598
QESTRAL (Part 4): Test Signals, Combining Metrics, and the Prediction of Overall Spatial Quality
Dewhirst, Martin; Conetta, Robert; Rumsey, Francis; Jackson, Philip; Zielinski, Slawomir; George, Sunish; Bech, Søren; Meares, David
The QESTRAL project has developed an artificial listener that compares the perceived quality of a spatial audio reproduction to a reference reproduction. Test signals designed to identify distortions in both the foreground and background audio streams are created for both the reference and the impaired reproduction systems. Metrics are calculated from these test signals and are then combined using a regression model to give a measure of the overall perceived spatial quality of the impaired reproduction compared to the reference reproduction. The results of the model are shown to match closely the results obtained in listening tests. Consequently, the model can be used as an alternative to listening tests when evaluating the perceived spatial quality of a given reproduction system, thus saving time and expense.

7599
An Unintrusive Objective Model for Predicting the Sensation of Envelopment Arising from Surround Sound Recordings
George, Sunish; Zielinski, Slawomir; Rumsey, Francis; Conetta, Robert; Dewhirst, Martin; Jackson, Philip; Meares, David; Bech, Søren
This paper describes the development of an unintrusive objective model, developed independently as a part of the QESTRAL project, for predicting the sensation of envelopment arising from commercially available 5-channel surround sound recordings. The model was calibrated using subjective scores obtained from listening tests that used a grading scale defined by audible anchors. For predicting subjective scores, a number of features based on Inter-aural Cross Correlation (IACC), Karhunen-Loeve Transform (KLT) and signal energy levels were extracted from recordings. The ridge regression technique was used to build the objective model and a calibrated model was validated using a listening test scores database obtained from a different group of listeners, stimuli and location. The initial results showed a high correlation between predicted and actual scores obtained from the listening tests.

7600
Accuracy Issues in Finite Element Simulation of Loudspeakers
Macey, Patrick
Finite element based software for simulating loudspeakers has been around for some time but is being used more widely now, due to improved solver functionality, faster hardware and improvements in links to CAD software and other preprocessing improvements. The analyst has choices to make in what techniques to employ, what approximations might be made and how much detail to model.

7601
Nonlinear Loudspeaker Unit Modeling
Pedersen, Bo Rohde; Agerkvist, Finn T.
Simulations of a 6½-inch loudspeaker unit are performed and compared with a displacement measurement. The non-linear loudspeaker model is based on the major nonlinear functions and expanded with time-varying suspension behaviour and flux modulation. The results are presented with FFT plots of three frequencies and different displacement levels. The model errors are discussed and analysed including a test with loudspeaker unit where the diaphragm is removed.

7602
An Optimized Pair-Wise Constant Power Panning Algorithm for Stable Lateral Sound Imagery in the 5.1 Reproduction System
Kim, Sungyoung; Ikeda, Masahiro; Takahashi, Akio
Auditory image control in the 5.1 reproduction system has been a challenge due to the arrangement of loudspeakers, especially in the lateral region. To suppress typical artifacts in pair-wise constant power algorithm, a new gain ratio between Left and Left Surround channel has been experimentally determined. Listeners were asked to estimate the gain ratio between two speakers for seven lateral positions so as to set the direction of the sound source. From these gain ratios, a polynomial function was derived of in order to that parametrically represented the gain ratio in an arbitrary direction was derived. The result of validating experiments showed that the new function produced stable auditory imagery in the lateral region.

7603
The Use of Delay Control for Stereophonic Audio Rendering Based on VBAP
Hyun, Dongil; Choi, Tacksung; Youn, Daehee; Lee, Seokpil; Park, Youngcheol
This paper proposes a new panning method that can enhance the performance of the stereophonic audio rendering system based on VBAP. The proposed system introduces a delay control to enhance the performance of the VBAP. Sample delaying is used to reduce the energy cancellation due to out-of-phase. Preliminary simulations and measurements are conducted to verify the controllability of ILD by delay control between stereophonic speakers. By simulating ILD by the delay control, spatial direction at frequencies where energy cancellation occurred could be perceived more stable than the conventional VBAP. The performance of the proposed system is also verified by subjective listening test.

7604
Ambience Sound Recording Utilizing Dual MS (Mid-Side) Microphone Systems Based upon Frequency Dependent Spatial Cross Correlation (FSCC)—Part-2: Acquisition of On-Stage Sounds
Muraoka, Teruo; Miura, Takahiro; Ifukube, Tohru
In musical sound recording, a forest of microphones is commonly observed. It is for good sound localization and favorable ambience, however the forest is desired to be sparse for less laborious setting up and mixing. For this purpose, the authors studied sound-image representation of stereophonic microphone arrangements utilizing Frequency Dependent Spatial Cross Correlation (FSCC), which is a cross correlation of two microphone's outputs. The authors firstly examined FSCCs of typical microphone arrangements for acquisition of ambient sounds and concluded that MS(Mid-Side) microphone system with setting directional azimuth at 132 degree is the best. The authors also studied conditions of on-stage sounds acquisition and resulted that FSCC of co-axial type microphone takes the constant value of +1, which is advantageous for stable sound localization. Thus the authors further compared additional sound acquisition characteristics of MS system (setting directional azimuth at 120 degree and XY system. As a conclusion, the former is superior. Finally, the author proposed dual MS microphone systems. One is for on-stage sound acquisition set directional azimuth at 120 degree and the other is for ambient sound acquisition set directional azimuth at 132 degree.

7605
Ambisonic Loudspeaker Arrays
Benjamin, Eric
The Ambisonic system is one of very few surround sound systems which offers the promise of reproducing full three-dimensional (periphonic) audio. It can be shown that arrays configured as regular polyhedra can allow the recreation of an accurate sound field at the center of the array. But the regular polyhedric shape can be impractical for real everyday usage because the requirement that the listener have his head located at the center of the array forces the location of the lower speakers to be beneath the floor, or even the location of a loudspeaker directly beneath the listener. This is obviously impracticable, especially in domestic applications. Likewise, it is typically the case that the width of the array is larger than can be accommodated within the room boundaries. The infeasibility of such arrays is a primary reason why they have not been more widely deployed. The intent of this work is to explore the efficacy of alternative array shapes for both horizontal and periphonic reproduction.

7606
Optimum Placement for Small Desktop/PC Loudspeakers
Filevski, Vladimir
A desktop/PC loudspeaker usually stands on a desk, so the direct sound from the loudspeaker interferes with the reflected sound from the desk. On the desk, a "perfect" loudspeaker with flat anechoic frequency response will not give a flat, but a comb-like resultant frequency response. Here is presented one simple and inexpensive solution to this problem - a small, conventional loudspeaker is placed on a holder. Holder is a horizontally pivoting telescopic arm, that enables easy positioning of the loudspeaker. With one side, the arm is attached on top corner of the PC monitor, and with the other side it is attached to the loudspeaker. The listener extends and rotates the arm in horizontal plane to such position that no reflection from the desk or from the PC monitor reaches the listener, thus preserving the presumably flat anechoic frequency response of the loudspeaker.

7607
An Audio Reproduction Grand Challenge: Design a System to Sonic Boom an Entire House
Sparrow, Victor W.; Garrett, Steven L.
This paper describes an ongoing research study to design a simulation device that can accurately reproduce sonic booms over the outside surface of an entire house. Sonic booms and previous attempts to reproduce them will be reviewed. The authors will present some calculations which suggest that it will be very difficult to produce the required pressure amplitudes using conventional sound reinforcement electroacoustic technologies. However, an additional purpose is to make AES members aware of this research and to solicit feedback from attendees prior to a January 2009 down-selection activity for the design of an outdoor sonic boom simulation system.

7608
A Platform for Audiovisual Telepresence Using Model- and Data-Based Wave-Field Synthesis
Heinrich, Gregor; Jung, Christoph; Hahn, Volker; Leitner, Michael
We present a platform for real-time transmission of immersive audiovisual impressions using model- and data-based audio wave-field analysis/synthesis and panoramic video capturing/projection. The audio subsystem considered in this paper is based on microphone arrays with different element counts and directivities as well as weakly directional loudspeaker arrays. We report on both linear and circular setups that feed different wave-field synthesis systems. In an attempt to extend this, we present first findings for a data-based approach derived using experimental simulations. This data-based wave-field analysis/synthesis (WFAS) approach uses a combination of cylindrical-harmonic decomposition of cardioid array signals and enforces causal plane wave synthesis by angular windowing and a directional delay term. Specifically, our contributions include (1) a high-resolution telepresence environment that is omnidirectional in both the auditory and visual modality, as well as (2) a study of data-based WFAS realistic microphone directivities as a contribution towards for real-time holophonic reproduction.

7609
SMART-I2: "Spatial Multi-User Audio-Visual Real Time Interactive Interface"
Marc, Rébillat; Corteel, Etienne; Katz, Brian F.
The SMART-I2 aims at creating a precise and coherent virtual environment by providing users with both audio and visual accurate localization cues. It is known that for audio rendering, Wave Field Synthesis, and for visual rendering, Tracked Stereoscopy, individually permit high quality spatial immersion within an extended space. The proposed system combines these two rendering approaches through the use of a large Multi-Actuator Panel used as both a loudspeaker array and as a projection screen, considerably reducing audio-visual incoherencies. The system performance has been confirmed by an objective validation of the audio interface and a perceptual evaluation of the audio-visual rendering.

7610
Head-Related Transfer Functions Reconstruction from Sparse Measurements Considering a Priori Knowledge from Database Analysis: A Pattern Recognition Approach
Guillon, Pierre; Nicol, Rozenn; Simon, Laurent
Individualized Head-Related Transfer Functions (HRTFs) are required to achieve high quality Virtual Auditory Spaces. This study proposes to decrease the total number of measured directions in order to make acoustic measurements more comfortable. To overcome the limit of sparseness for which classical interpolation techniques fail to properly reconstruct HRTFs, additional knowledge has to be injected. Focusing on the spatial structure of HRTFs, the analysis of a large HRTF database enables to introduce spatial prototypes. After a pattern recognition process, these prototypes serve as a well-informed background for the reconstruction of any sparsely measured set of individual HRTFs. This technique shows better spatial fidelity than blind interpolation techniques.

7611
Near-Field Compensation for HRTF Processing
Romblom, David; Cook, Bryan
It is difficult to present near-field virtual audio displays using available HRTF filters, as most existing databases are measured at a single distance in the far-field of the listener’s head. Measuring near-field data is possible, but would quickly become tiresome due to the large number of distances required to simulate sources moving close to the head. For applications requiring a compelling near-field virtual audio display, one could compensate the far-field HRTF filters with a scheme based on 1/r spreading roll off. However, this would not account for spectral differences that occur in the near-field. Using difference filters based on a spherical head model [1], as well as a geometrically accurate HRTF lookup scheme [2], we are able to compensate existing data and present a convincing virtual audio display for near field distances.

7612
A Method for Estimating Interaural Time Difference for Binaural Synthesis
Nam, Juhan; Abel, Jonathan S.; Smith III, Julius O.
A method for estimating interaural time difference (ITD) from measured head-related transfer functions (HRTFs) is presented. The method forms ITD as the difference in left-ear and right-ear arrival times, estimated as the times of maximum cross-correlation between measured HRTFs and their minimum-phase counterparts. This arrival time estimate is related to a nonlinear least-squares fit to the measured excess phase, emphasizing those frequencies having large HRTF magnitude and deweighting large phase delay errors. As HRTFs are nearly minimum-phase, this method is robust compared to the conventional approach of cross-correlating left-ear and right-ear HRTFs, which can be very different. The method also performs slightly better than techniques averaging phase delay over a limited frequency range.

7613
Efficient Delay Interpolation for Wave Field Synthesis
Franck, Andreas; Brandenburg, Karlheinz; Richter, Ulf
Wave Field Synthesis enables the reproduction of complex auditory scenes and moving sound sources. Moving sound sources induce time-variant delay of source signals. To avoid severe distortions, sophisticated delay interpolation techniques must be applied. The typically large numbers of both virtual sources and loudspeakers in a WFS system result in a very high number of simultaneous delay operations, thus being a most performance-critical aspect in a WFS rendering system. In this article, we investigate suitable delay interpolation algorithms for WFS. To overcome the prohibitive computational cost induced by high-quality algorithms, we propose a computational structure that achieves a significant complexity reduction through a novel algorithm partitioning and efficient data reuse.

7614
Obtaining Binaural Room Impulse Responses from B-Format Impulse Responses
Menzer, Fritz; Faller, Christof
Given a set of head related transfer functions (HRTFs) and a room impulse response measured with a Soundfield microphone, the proposed technique computes binaural room impulse responses (BRIRs) which are similar to binaural room impulse responses that would be measured if in place of the Soundfield micro- phone, the dummy head used for the HRTF set was directly recording the BRIRs. The proposed technique enables that from a set of HRTFs corresponding BRIRs for different rooms are obtained without a need for the dummy head or person to be present for measurement.

7615
A New Audio Postproduction Tool for Speech Dereverberation
Kinoshita, Keisuke; Nakatani, Tomohiro; Miyoshi, Masato; Kubota, Toshiyuki
This paper proposes a new audio post-production tool for speech dereverberation that utilizes our previously proposed method. In previous studies, we proposed a single-channel dereverberation method as preprocessing of automatic speech recognition and reported its good performance. This paper focuses more on the improvement of the audible quality of the dereverberated signals. To achieve good dereverberation with less audible artifacts, the previously proposed dereverberation method is combined with post-processing that implicitly considers the perceptual masking property. The system has three adjustable parameters for controlling audible quality. With an informal evaluation, we found that the proposed tool allows the professional audio engineers to dereverberate a set of reverberant recordings efficiently.

7616
Preliminary Results of Calculation of a Sound Field Distribution for the Design of a Sound Field Effector Using a 2-Way Loudspeaker Array with Pseudorandom Configuration
Iijima, Yoshihiro; Ashihara, Kaoru; Kiryu, Shogo
We have been developing a loudspeaker array system which can be controlled sound field in real time for live concerts. In order to reduce the sidelobes and to improve the frequency range, a 2-way loudspeaker array with pseudorandom configuration is proposed. A software has been developing to determine the configuration. For now, the configuration is optimized for a focused sound. The software calculates the ratio between the sound pressure of the focus point and the average of the sound pressure around the focus. It was shown that the sidelobe can be reduced with a pseudorandom configuration.

7617
Design and Implementation of a Sound Field Effector Using a Loudspeaker Array
Hayashi, Seigo; Tanno, Tomoaki; Kamekawa, Toru; Ashihara, Kaoru; Kiryu, Shogo
We have been developing an effector which uses a 128-channel two-way loudspeaker array system for live concerts. The system was designed to ealize the change of the sound field within 10 ms. The variable delay circuits and the communication circuit between the hardware and the control computer are implemented in one FPGA. All of the delay data which have been calculated in advance are stored in the SDRAM which is mounted on the FPGA board and only the simple command is sent from the control computer. The system can control up to four sound focuses independently.

7618
Wave Field Synthesis: Practical Implementation and Application to Sound Beam Digital Pointing
Peretti, Paolo; Romoli, Laura; Palestini, Lorenzo; Cecchi, Stefania; Piazza, Francesco
Wave Field Synthesis (WFS) is a digital signal processing technique introduced to achieve an optimal acoustic sensation in a larger area than in traditional systems (Stereophony, Dolby Digital). It is based on a large number of loudspeakers and its real-time implementation needs the study of efficient solutions in order to limit the computational cost. To this end, in this paper we propose an approach based on a preprocessing of the driving function component, which does not depend on the audio streaming. Linear and circular geometries tests will be described and the application of this technique to digital pointing of the sound beam will be presented.

7619
Highly Focused Sound Beamforming Algorithm Using Loudspeaker Array System
Hur, Yoomi; Kim, Seong Woo; Park, Young-cheol; Youn, Dae Hee
This paper presents a technique of sound beamforming using a loudspeaker array that can generate highly focused sound beams. First, we find an optimal solution to the problem of maximizing the sound power ratio between a target region and the total region excluding the target region. A preliminary sound beam is designed using the optimal weight. Next, to enhance the directivity of the pre-designed sound beam, an iterative pattern synthesis technique, which was introduced for antenna array, is applied. By assuming that there are imaginary sound powers in the non-target regions, the algorithm can iteratively improve the directivity of the sound beam. The performance of proposed method was evaluated, and the results showed that it could provide highly focused sound beam compared to the conventional method.

7620
Super-Directive Loudspeaker Array for the Generation of Personal Sound Zone
Choi, Jung-Woo; Kim, Youngtae; Ko, Sangchul; Kim, Jung-Ho
A sound manipulation technique is proposed for selectively enhancing a desired acoustic property in a zone of interest called personal sound zone. In order to create the personal sound zone in which a listener can experience high sound level, acoustic energy is focused on only a selected area. Recently, two performance measures indicating acoustic properties of the personal sound zone –acoustic brightness and contrast – are employed to optimize driving functions of a loudspeaker array. In this paper, firstly some limitations of individual control method are presented, and then a novel control strategy is suggested such that advantages of both are combined in a single objective function. Precise control of sound field with desired shape of energy distribution is made possible by introducing continuous spatial weighting technique. The results are compared to those based on the least-square optimization technique.

7621
A Framework for a Near-Optimal Excitation Based Rate-Distortion Algorithm for Audio Coding
Vilermo, Miikka
An optimal excitation based rate-distortion algorithm remains an elusive target in audio coding. Typical complexity of the problem for one frame alone is in the order of 60^50. This paper presents a framework for reducing that complexity. Excitation is calculated using cochlear filters that have relatively steep slopes above and below the central frequency of the filter. An approximation of the excitation can be calculated by limiting the cochlear filters to a small frequency region. For example, the cochlear filters may span 15 subbands. In this way, the complexity can be reduced approximately to the order of 60^15*50.

7622
Audio Bandwidth Extension by Frequency Scaling of Sinusoidal Partials
Zernicki, Tomasz; Bartkowiak, Maciej
This paper describes a new technique of efficient coding of high-frequency signal components as an alternative to Spectral Band Replication. The main idea is to reconstruct the high frequency harmonic structure trajectories by using fundamental frequencies obtained at the encoder side. Audio signal is decomposed into narrow subbands by demodulation based on the local instantaneous frequency of individual partials. High frequency components are reconstructed by modulation of the baseband signals with appropriately scaled instantaneous frequencies. Such approach offers correct synthesis of rapidly changing sinusoids as well as proper reconstruction of harmonic structure in the high-frequency band. This technique performs also a correct energy adjustment of sinusoidal partials. High compression efficiency has been achieved and confirmed by listening tests.

7623
Robustness Issues in Multi-View Audio Coding
Väänänen, Mauri
This paper studies the problem of noise unmasking when multiple spatial filtering options (multiple views) are required from multi-microphone recordings compressed with lossy coding. The envisaged application is re-use and post-processing of user-created content. A potential solution based on inter-channel prediction is outlined, that would allow also subtractive downmix options without excessive noise unmasking. The simple case of two relatively closely spaced omni-directional microphones and mono downmix is used as an example, experimenting with real-world recordings and MPEG-1 Layer 3 coding.

7624
Quality Improvement of Very Low Bit Rate HE-AAC Using Linear Prediction Module
Lee, GunWoo; Lee, JaeSeong; Park, YoungCheol; Youn, DaeHee
This paper proposes a new method of improving the quality of High Efficiency Advanced Audio Coding (HE-AAC) at very low bitrate under 16kbps. Low bitrate HE-AAC often produces obvious spectral holes inducing musical noise in low energy frequency bands, due to limited number of available bits. In the proposed system, a Linear Prediction (LP) module is combined with HE-AAC as a pre-processor to reduce the spectral holes. In its implementation, masking threshold of psychoacoustic model is normalized using the LP spectral envelope in prior to quantization of the LP residual. Also, in order to reduce the pre-echo, a block switching module is modified. Experimental results show that, at very low bitrate modes, the linear prediction module effectively reduces spectral holes, which results in reduction of musical noises of the conventional HE-AAC.

7625
An Implementation of MPEG-4 ALS Standard Compliant Decoder on ARM Core CPUs
Harada, Noboru; Moriya, Takehiro; Kamamoto, Yutaka
MPEG-4 Audio Lossless Coding (ALS) is a standard that losslessly compresses audio signals in an efficient manner. MPEG-4 ALS is a suitable compression scheme for high-sound-quality portable music players. We have implemented a decoderder compliant with the MPEG-4 ALS standard on the ARM platform. In this paper, the required CPU resources for MPEG-4 ALS tools on ARM9E are characterized by using an ARM CPU emulator, called ARMulator, as a simulation platform. It is shown that the required CPU clock cycle for decoding MPEG-4 ALS standard compliant bitstreams is less than 20 MHz for 44.1-kHz-16-bit, stereo signals on ARM9E when the combination of the MPEG-4 ALS tools is properly selected and coding parameters are properly restricted.

7626
Assessing the Acoustic Performance and Potential Intelligibility of Assistive Audio Systems for the Hard of Hearing and Other Users
Mapp, Peter
Around 14% of the general population suffer from a noticeable degree of hearing loss and would benefit from some form of hearing assistance or deaf aid. Recent DDA legislation and requirements mean that many more hearing assistive systems are being installed – yet there is evidence to suggest that many of these systems fail to perform adequately and provide the benefit expected. There has also been a proliferation of classroom and lecture room “soundfield” systems, with much conflicting evidence as to their apparent effectiveness. This paper reports on the results of some trial acoustic performance testing of such systems. In particular the effects of system microphone type, distance and location are shown to have a significant effect on the resultant performance. The potential of using the Sound Transmission Index (STI) and in particular STIPa, for carrying out installation surveys has been investigated and a number of practical problems are highlighted. The requirements for a suitable acoustic test source to mimic a human talker are discussed as is the need to the need to adequately assess the effects of both reverberation and noise. The findings discussed in the paper are also relevant to the installation and testing of boardroom and conference room telecommunication systems.

7627
Aging and Sound Perception: Desirable Characteristics of Entertainment Audio for the Elderly
Müsch, Hannes
During the last few years the research community has made substantial progress towards understanding how aging affects the way the ear and brain process sound. A review of the literature supports our experience as audio professionals that elderly listeners have preferences for the reproduction of entertainment audio that differ from those of young listeners. This presentation reviews the literature on aging and sound perception with a focus on speech. The review identifies desirable audio reproduction characteristics and discusses signal processing techniques to generate audio that is suited for elderly listeners.

7628
Speech Enhancement of Movie Sound
Uhle, Christian; Hellmuth, Oliver; Weigel, Jan
Today, many people have problems understanding the speech content of a movie, e.g. due to hearing impairments. This paper describes a method for improving the speech intelligibility of movie sound. Speech is detected by means of a pattern recognition method; the audio signal is then attenuated during periods where speech is absent. The speech signals are further processed by a spectral weighting method aiming at the suppression of the background noise. The spectral weights are computed by means of feature extraction and a neural network regression method. The output signal finally carries all relevant speech with reduced background noise allowing the listener to follow the plot of the movie more easily. Results of numerical evaluations and of listening tests are presented.

7629
An Investigation of Audio Balance for Elderly Listeners Using Loudness as the Main Parameter
Komori, Tomoyasu; Takagi, Tohru; Kurozumi, Kohichi; Murakawa, Kazuhiro
We have been studying the best sound balance for audibility for elderly listeners. We conducted subjective tests on the balance between narration and background sound using professional sound mixing engineers. The comparative loudness of narration to background sound was used to calculate appropriate respective levels for use in documentary programs. Monosyllabic intelligibility tests were then conducted in a noisy environment with both elderly and young people and a condition that complicates hearing for the elderly was identified. Assuming that the recruitment phenomenon and reduced ability to separate narration from background sound cause hearing problems for the elderly, we estimated appropriate loudness levels for them. We also constructed a prototype to assess the best audio balance for the elderly objectively.

7630
Estimating the Transfer Function from Air Conduction Recording to One’s Own Hearing
Won, Sook Young; Berger, Jonathan
It is well known that there is often a sense of disappointment when an individual hears a recording of his/her own voice. The perceptual disparity between the live and recorded sound of one’s own voice can be explained scientifically as the result of the multiple paths via which our body transmits vibrations from the vocal cords to the auditory system during vocalization, as opposed to the single air-conducted path in hearing a playback of one’s own recorded voice. In this paper, we aim to investigate the spectral characteristics of one’s own hearing as compared to an air-conducted recording. To accomplish this objective, we designed and conducted a perceptual experiment with a real-time filtering application.

7631
Determination and Correction of Individual Channel Time Offsets for Signals Involved in an Audio Mixture
Perez Gonzalez, Enrique; Reiss, Joshua
A method for reducing comb-filtering effects due to delay time differences between audio signals in a sound mixer has been implemented. The method uses a multi-channel cross-adaptive effect topology to automatically determine the minimal delay and polarity contributions required to optimize the sound mixture. The system uses real time, time domain transfer function measurements to determine and correct the individual channel offset for every signal involved in the audio mixture. The method has applications in live and recorded audio mixing where recording a single sound source with more than one signal path is required, for example when recording a piano with multiple microphones. Results are reported which determine the effectiveness of the proposed method.

7632
STFT-Domain Estimation of Subband Correlations
Goodwin, Michael M.
Various frequency-domain and subband audio processing algorithms for upmix, format conversion, spatial coding, and other applications have been described in the recent literature. Many of these algorithms rely on measures of the subband autocorrelations and cross-correlations of the input audio channels. In this paper we consider several approaches for estimating subband correlations based on a short-time Fourier transform representation of the input signals. We discuss application of the estimation methods to ambience extraction and address several practical implementation issues.

7633
Separation of Singing Voice from Music Accompaniment with Unvoiced Sounds Reconstruction for Monaural Recordings
Hsu, Chao-Ling; Jang, Jyh-Shing Roger; Tsai, Te-Lu
Separating singing voice from music accompaniment is an appealing but challenging problem, especially in the monaural case. One existing approach is based on computational audio scene analysis which uses pitch as the cue to resynthesize the singing voice. However, the unvoiced parts of the singing voice are totally ignored since they have no pitch at all. This paper proposes a method to detect unvoiced parts of an input signal and to resynthesize them without using pitch information. The experimental result demonstrates that the unvoiced parts can be reconstructed successfully, with 3.28 dB signal-to-noise ratio higher than that achieved by the currently state-of-the-art method in the literature.

7634
Low Latency Convolution in One Dimension Via Two Dimensional Convolutions: An Intuitive Approach
Hurchalla, Jeffrey
This paper presents a class of algorithms which can be used to efficiently perform the running convolution of a digital signal with a finite impulse response. The impulse is uniformly partitioned and transformed into the frequency domain, changing the one dimensional convolution into a two dimensional convolution which can be efficiently solved with nested short length acyclic convolution algorithms applied in the frequency domain. The latency of the running convolution is the time needed to acquire a block of data equal in size to the uniform partition length.

7635
Simple Arbitary IIRs
Lee, Richard
This is a method of fitting IIRs (Infinite Impulse Response filters) to an arbitrary frequency response, simple enough to incorporate in intelligent AV receivers. Short IIR filters are useful where computational power is limited and at low frequencies where FIRs have poor performance. Speaker and microphone frequency response defects are often better matched to IIRs. Some caveats for digital EQ design are discussed. The emphasis is on speakers and microphones.

7636
Analysis of Design Parameters for Crosstalk Cancellation Filters Applied to Different Loudspeaker Configurations
Lacouture Parodi, Yesenia
Several approaches to render binaural signals through loudspeakers have been proposed in past decades. Some studies have focused on the optimum loudspeaker arrangement while others have proposed more efficient filters. However, to our knowledge, the identification of optimal parameters for crosstalk cancellation filters applied to different loudspeakers configurations has not yet been addressed systematically. In this paper, we document a study of three different inversion techniques applied to several loudspeaker arrangements. Least square approximations in frequency and time domain are evaluated along with a crosstalk canceler based on minimum-phase approximation. The three methods are simulated in two-channel configuration and the least square approaches in four-channel configurations. Different span angles and elevations are evaluated for each case. In order to obtain optimum parameter, we varied the bandwidth, filter length and regularization constant for each loudspeaker position and each method. We present a description of the simulations carried out and the optimum regularization values, expected channel separation and performance error obtained for each configuration.

7637
A Hybrid Time and Frequency Domain Audio Pitch Shifting Algorithm
Juillerat, Nicolas; Arisona, Stefan Müller; Schubiger-Banz, Simon
This paper presents an abstract algorithm that performs audio pitch shifting as a combination of a signal analysis, a filter bank and frequency shifting operations. Then, it is shown that two previously proposed pitch shifting techniques are actually concrete implementations of the presented abstract algorithm. One of them is implemented in the frequency domain whereas the other is implemented in the time domain. Based on an analysis and comparison of the properties of these two techniques (quality, artifacts, assumptions on the signal), we propose a new hybrid implementation working partially in the frequency domain and partially in the time domain, and achieving superior quality by taking the best from each of the two existing techniques.

7638
A Colored Noise Suppressor Using Lattice Filter with Correlation Controlled Algorithm
Kawamura, Arata; Iiguni, Youji
A noise suppression technique is necessary in a wide range of applications including mobile communication and speech recogunition systems. We have previously proposed a noise suppressor using lattice filter which can cancel a white noise from an observed signal. Unfortunately, many practical noises are not white, and hence the conventional noise suppressor is not available for the practical noises. In this paper, we propose a new adaptive algorithm used for the lattice filter to suppress a colored noise. The proposed algorithm can be directly derived from the conventional time recursive algorithm. To extract a speech from a speech mixed with colored noise, the lattice filter with the proposed algorithm gives a noise replica whose auto-correlation is close to the noise's one. Subtracting the noise replica from the observed noisy speech, we can obtain an extracted speech. Simulation results showed that the proposed noise suppressor can extract a speech from a speech mixed with a tunnel noise which is a colored noise recorded in a practical environment.

7639
Accurate IIR Equalization to an Arbitrary Frequency Response, with Low Delay and Low Noise Real-Time Adjustment
Eastty, Peter
A new form of equalizer has been developed which combines minimum phase, low delay, IIR signal processing with low noise, real-time adjustment of coefficients to accurately deliver an arbitrary frequency response as entered from a graphical user interface. The use of a join-the-dots type graphical user interface combined with cubic or similar splines is a common method of entering curved lines into 2D drawing programs. The equalizer described in this paper combines a similar type of user interface with low-delay, minimum phase, IIR audio DSP. Key attributes also include real-time, nearly noiseless adjustment of the DSP coefficients in response to user input. A demonstration and all necessary information for the construction of these filters is included.

7640
A Method of Capacity Increase for Time-Domain Audio Watermarking Based on Low-Frequency Amplitude Modification
Murata, Harumi; Ogihara, Akio; Iwata, Motoi; Shiozaki, Akira
The objective of this work is to increase the capacity of watermark information in “the audio watermarking method based on amplitude modification” which has been proposed by W.N.Lie as a prevention technique against the copyright infringement. In this conventional method, the capacity of watermark information is not enough, and it is desirable that the capacity of watermark information is increased. In this paper, we increase the capacity of watermark information by embedding multiple watermarks in the different levels of audio data independently. The proposed method has many data-channels for embedding, and hence it is possible to embed multiple watermarks by selecting proper data-channel according to required data capacity or recovery rate.

7641
Constrained-Optimized Sound Beamforming of Loudspeaker-Array System
Song, Myung; Baek, Soon-Ho; Lee, Seok-Pil; Kang, Hong-Goo
This paper proposes a novel speaker-array system to form relatively high sound pressure toward the desired location. The proposed algorithm adopts a constrained-optimization technique such that the array response to the desired response is maintained over mainlobe width while minimizing its sidelobe level. At first the characteristic of sound propagation in reverberant environment is analyzed by off-line computer simulation. Then, the performance of implemented speaker-array system is evaluated by measuring sound pressure distribution in a real test room. The results show that the proposed sound beamforming algorithm forms more concentrative sound beam to desired location than conventional algorithms even in reverberation environment.

7642
Magnetic Development: Magneto-Optical Indicator Film Imaging vs. Ferrofluids
Broyles, Jonathan C.
Techniques, advantages and disadvantages of ferrofluids and magneto-optical indicator film imaging methods of magnetic development are discussed. Presentation and discussion of test results with supporting test images and figures. A number of MOIF imaging examples are presented for discussion. General overview on how magneto-optical magnetic development works. Magnetic development examples processed on magneto-optical imaging system developed and built by the author.

7643
Extraction of Electric Network Frequency Signals from Recordings Made in a Controlled Magnetic Field
Sanders, Richard; Popolo, Peter S.
An Electric Network Frequency (ENF) signal is the 60 Hz component of an AC power signal that varies over time due to fluctuations in power production and consumption, across the entire grid of a power distribution network. When present in audio recordings, these signals (or their harmonics) can be used to authenticate a recording, time stamp the original, or determine if a recording was copied or edited. This paper will present the results of an experiment to determine if ENF signals in a controlled magnetic field can be detected and extracted from audio recordings made with battery and externally powered audio recording devices.

7644
Forensic Voice Identification Utilizing Digitally Extracted Speech Characteristics
Smith, Jeff M.; Sanders, Richard
By combining modern capabilities in the digital domain with more traditional methods of aural comparison and spectrographic inspection, the acquisition of identity from recorded evidence can be executed with greater confidence. To aid the Audio Forensic Examiner’s efforts in this, an effective approach to manual voice identification is presented here. To this end, this paper relates the research into and application of unique vocal characteristics utilized by the SIDNI (Speaker Identification by Numerical Imprint) automated system of voice identification to manual forensic investigation. Some characteristics include: fundamental speaking frequency, rate of vowels, proportional relationships in spectral distribution, amplitude of speech, and perturbation measurements.

7645
Applications of Algorithmically-Generated Digital Audio for Web-Based Sonic Measure Ear Training
Ariza, Christopher
This paper examines applications of algorithmically-generated digital audio for a new type of ear training. This approach, called sonic measure ear training, circumvents the many limits of MIDI-based aural testing, and may offer a valuable resource for computer musicians and audio engineers. The Post-Ut system, introduced here, is the first web-based ear training system to offer sonic measure ear-training. After describing the design of the Post-Ut system, including the use of athenaCL, Csound, Python, and MySQL, the audio generation procedures are examined in detail. The design of questions and perceptual considerations are evaluated, and practical applications and opportunities for future development are outlined.

7646
A Perceptual Model-Based Speech Enhancement Algorithm
Yu, Rongshan
This paper presents a perceptual model based speech enhancement algorithm. The proposed algorithm measures the amount of the audible noise in the input noisy speech explicitly by using a psychoacoustic model, and decides an appropriate amount of noise reduction accordingly to achieve good noise level reduction without introducing significant distortion to the clean speech embedded in the input noisy signal. The proposed algorithm also mitigates the musical noise problem commonly encountered in conventional speech enhancement algorithms by having the amount of noise reduction adapt to the instantly estimated noise amplitude. Good performance of the proposed algorithm has been confirmed through objective and subjective tests.

7647
Real Time Implementation of an ESPRIT-Based Bass Enhancement Algorithm
Palestini, Lorenzo; Moretti, Emanuele; Peretti, Paolo; Cecchi, Stefania; Romoli, Laura; Piazza, Francesco
This paper presents a software real-time implementation for the NU-Tech platform of a bass enhancement algorithm based on the FAPI subspace tracker and the ESPRIT algorithm for fundamentals estimation to realize bass improvement of small loudspeakers exploiting the well known psychoacoustic phenomenon of the missing fundamental. Comparative informal listening tests have been performed to validate the virtual bass improvement and their results show that the proposed method is well appreciated.

7648
Low-Power Implementation of a Subband Acoustic Echo Canceller for Portable Devices
Johnson, Julie; Hermann, David; Wdowiak, John; Chau, Edward; Sheikhzadeh, Hamid
Portable audio communication devices require increasingly superior audio quality while using minimal power. Devices such as cell phones with speakerphone functionality can generate substantial acoustic echo due to the proximity of the microphone and speaker. To improve the audio quality in such devices, an oversampled subband acoustic echo canceller has been implemented on a miniature low-power dual-core DSP system. This application is comprised of three subband-based algorithms: a Pseudo-Affine Projection adaptive filter, an Ephraim-Malah-based single-microphone noise reduction algorithm and a novel non-linear residual echo suppressor. The system consumes less than 4 mW of power when configured with a 128 ms filter. Real-world tests indicate an echo return loss enhancement of greater than 30 dB for typical input levels.

7649
A Digital Model of the Echoplex Tape Delay
Arnardottir, Steinunn; Abel, Jonathan S.; Smith III, Julius O.
The Echoplex is a tape delay unit featuring fixed playback and erase heads, a movable record head, and a tape loop moving at roughly 8 ips. The relatively slow tape speed allows large frequency shifts, including "sonic booms" and shifting of the tape bias signal into the audio band. Here, the Echoplex tape delay is modeled with read, write and erase pointers moving along a circular buffer. The model separately generates the quasiperiodic capstan and pinch wheel components and drift of the observed fluctuating time delay. This delay drives an interpolated write simulating the record head. To prevent aliasing in the presence of a changing record head position, an anti-aliasing filter with a variable cutoff frequency is described.

7650
A Digital Reverberator Modeled After the Scattering of Acoustic Waves by Trees in a Forest
Spratt, Kyle; Abel, Jonathan S.
A digital reverberator modeled after the scattering of acoustic waves among trees in an idealized forest is presented. Termed ``treeverb,' the technique simulates forest acoustics using a network of digital waveguides, with bi-directional delay lines connecting trees represented by multi-port scattering junctions. The reverberator is designed by selecting tree locations and diameters, with waveguide delays determined by inter-tree distances, and scattering filters fixed according to tree-to-tree angles and trunk diameters. The scattering is modeled as that of plane waves normally incident on a rigid cylinder, and a simple low-order scattering filter model is presented. Small forests are seen to yield dense, gated reverb-like impulse responses.