Monday, May 22, 15:00 — 18:00
P19-01 Audio Time Stretching with Controllable Phase Coherence
Nicolas Juillerat (Presenting Author)
This paper presents a hybrid audio time stretching technique in which the trade-off between vertical and horizontal phase coherence can be freely controlled by a single parameter. Depending on that parameter, the proposed technique sounds like a time domain technique at one extreme, like a phase-locked vocoder at the other extreme, or anywhere in between. By properly choosing the value of the control parameter, it is possible to manually adjust the algorithm to the characteristics of the audio signal being transformed in order to get an optimal result. Furthermore, appropriate middle values yield good results for a wide range of audio signals with mixed content.
Convention Paper 9780
P19-02 Modelling Nonlinearities on Musical Instruments by Means of Volterra Series
Vanna Lisa Coli (Presenting Author), Francesco F. Gionfalo (Author), Lamberto Tronchin (Author)
The behavior of the soundboard of electroacoustic tools and musical instruments has being investigated for several years. The modelling of such instruments is fundamental in order to determine their acoustic characterization. The determination of nonlinear features of the sound production and propagation allows the definition of acoustical aspects that can’t be reproduced with methods based on linear impulse response. A method that allows approximating nonlinear distortions of musical instruments by exploiting the Volterra series model is presented. A Matlab code has been developed in order to test the method on real world audio signals. Results of applications are presented on a series of different wind instruments. Some sound examples are provided.
Convention Paper 9781
P19-03 The Influence of Source Spectrum and Loudspeaker Azimuth on Vertical Amplitude Panning
Maksims Mironovs (Presenting Author), Hyunkook Lee (Author)
Listening tests were conducted to examine the influence of source spectrum and loudspeaker azimuth on the accuracy of vertical amplitude panning. Subjects judged the perceived elevation of the phantom images created using vertical loudspeaker pairs placed at 0°and 30° azimuths. Six sound sources with different spectral characteristics were used: broadband, low-passed and high-passed pink noises as well as speech, bird and tank shot recordings. Results generally indicated that the localization accuracy was poor, however, lower or upper response biases observed in the results were found to be significantly dependent on the target panning angle, the stimuli and the loudspeaker azimuth angle. In particular, the low-passed noise presented from the loudspeakers at 30° azimuth was perceived to be significantly elevated.
Convention Paper 9782
P19-04 Efficient Natural Sample Calculation for Digital Pulse Width Modulation
Carsten Wegner (Presenting Author), Dietmar Ehrhardt (Author), Robert Schwann (Author)
In this paper, an improved algorithm for natural sampling is presented that is suitable for digitally controlled fixed frequency PWM modulators. With only 5 MAC operations, 4 multiplications, and 2 additions, the algorithm calculates both switching times for double-sided 3-level PWM, and offers more than 100 dB between signal and PWM related distortion products for high fidelity audio applications. These features compare well with results published [2-6]. The algorithm can be combined with a noise shaped local feedback for quantized pulse lengths and the digital modulator can be integrated into a global feedback loop.
Convention Paper 9783
P19-05 Construction of Lightweight Loudspeaker Enclosures
Herle Bagh Juul-Nyholm (Presenting Author), Michael A. E. Andersen (Author), Niels Henrik Mortensen (Author), Henrik Schneider (Author), Jonas Corfitz Severinsen (Author)
On the basis of bass cabinets, this paper deals with the problem of reducing loudspeaker enclosure weight. An introductory market analysis emphasizes that lighter cabinets are sought, but maintenance of sound quality is vital. The problem is challenged through experiments and simulations in COMSOL Multiphysics, which indicate that weight reduction and sound quality maintenance is possible by reducing wall thickness and using adequate bracing and lining.
Convention Paper 9784
P19-06 LAMI: A Gesturally Controlled Three-Dimensional Stage Leap (Motion-Based) Audio Mixing Interface
Jonathan Wakefield (Presenting Author), Christopher Dewey (Author), William Gale (Author)
Interface designers are increasingly exploring alternative approaches to user input/control. LAMI is a Leap (Motion-based) AMI that takes user’s hand gestures and maps these to a three-dimensional stage displayed on a computer monitor. Audio channels are visualized as spheres whose Y coordinate is spectral centroid and X and Z coordinates are controlled by hand position and represent pan and level respectively. Auxiliary send levels are controlled via wrist rotation and vertical hand position and visually represented as dial-like arcs. Channel EQ curve is controlled by manipulating a lathed column visualization. Design of LAMI followed an iterative design cycle with candidate interfaces rapidly prototyped, evaluated, and refined. LAMI was evaluated against Logic Pro X in a defined audio mixing task.
Convention Paper 9785
P19-07 OSPW (Open Signal Processing Workstation)–Development of a Stand-Alone Open Platform for Signal-Processing in AV-Networks
Holger Stenschke (Presenting Author), Clemens Fiechter (Author), Peter Glaettli (Author), Thomas Resch (Author), Roman Riedl (Author)
This paper presents the concept and design of a newly developed stand-alone, fully programmable signal processing platform for networked audio and music applications. In recognition of one of the first successful music DSP computation platforms, the ISPW , this prototype was named OSPW | Open Signal Processing Workstation. The first part of this paper describes the project's main objectives. The second part provides an overview of the OSPW system components, along with the technologies in use. The third part outlines proof-of-concept demo applications and gives an outlook as to potential user scenarios.
Convention Paper 9786
P19-08 Extending Temporal Feature Integration for Semantic Audio Analysis
Lazaros Vrysis (Presenting Author), Charalampos Dimoulas (Author), George Papanikolaou (Author), Nikolaos Tsipas (Author)
Semantic audio analysis has become a fundamental task in contemporary audio applications; consequently, further improvement and optimization of classification algorithms has also become a necessity. During the recent years, standard frame-based audio classification methods have been optimized and modern approaches introduced additional feature engineering steps, attempting to capture the temporal dependency between successive feature observations. This type of processing is known as Temporal Feature Integration. In this paper, the enhancement of statistical feature integration is proposed by extending and extensively evaluating the measures that can be deployed. Under this scope, new functions for capturing the shape of a texture window are introduced and evaluated. The ultimate goal of this work is to highlight the best performing measures for early temporal integration, focusing on simple feature engineering, avoiding complexity, and forming a compact and robust set of meta-features that can improve performance in audio classification tasks.
Convention Paper 9808