Last Updated: 20050819, wtm
Sunday, October 9, 1:00 pm — 2:30 pm
P15-1 The Design of Half-Band FIR Filters Using Ripple Attenuation of a Manipulated Lowpass—Duane Wise, Consultant - Boulder, CO, USA
This paper investigates a technique for extracting a half-band FIR filter from a lowpass root FIR. Employing this technique with a root filter designed with an optimal least squares algorithm can result in a half-band FIR with very low ripple over most of the pass-band and stop-band, at the expense of ripple size at the band-edges. The principal advantage of this technique is in the design of half-band FIRs with less priority to band-edge ripple without the manipulation of an arbitrary weighting function.
Convention Paper 6602 (Purchase now)
P15-2 Single Channel Source Separation Using Short-Time Independent Component Analysis—Dan Barry, Dublin Institute of Technology - Dublin, Ireland; Derry Fitzgerald, Cork Institute of Technology - Cork, Ireland; Eugene Coyle, Dublin Institute of Technology - Dublin, Ireland; Bob Lawlor, National University of Ireland - Maynooth, Ireland
In this paper we develop a method for the sound source separation of single channel mixtures using Independent Component Analysis within a time-frequency representation of the audio signal. We apply standard Independent Component Analysis techniques to contiguous magnitude frames of the short-time Fourier transform of the mixture. Provided that the amplitude envelopes of each source are sufficiently different, it can be seen that it is possible to recover the independent short-time power spectra of each source. A simple scoring scheme based on auditory scene analysis cues is then used to overcome the source ordering problem ultimately allowing each of the independent spectra to be assigned to the correct output source. A final stage of adaptive filtering is then applied, which forces each of the spectra to become more independent. Each of the sources is then resynthesized using the standard inverse short-time Fourier transform with an overlap add scheme.
Convention Paper 6603 (Purchase now)
P15-3 A New Class of Smooth Power Complementary Windows and their Application to Audio Signal Processing—Deepen Sinha, ATC Labs - Chatham, NJ, USA; Anibal Ferreira, University of Porto - Porto, Portugal and ATC Labs, Chatham, NJ, USA
In this paper we describe a new family of smooth power complementary windows, which exhibit a very high level of localization in both time and frequency domain. This window family is parameterized by a "smoothness quotient." As the smoothness quotient increases the window becomes increasingly localized in time (most of the energy gets concentrated in the center half of the window) and frequency (far field rejection becomes increasing small to the order of -150 dB or lower). A closed form solution for such window function exists and the associated design procedure is described. The new class of windows are quite attractive for a number of applications as switching functions, equalization functions, or as windows for overlap-add and modulated filterbanks. An extension to the family of smooth windows that exhibits improved near-field response in frequency domain is also described.
Convention Paper 6604 (Purchase now)
P15-4 Active Leak Compensation in Small-Sized Loudspeakers Using Digital Signal Processing—Varun Chopra, Chalmers University of Technology - Gothenburg, Sweden
The frequency response of a small-sized loudspeaker unit used in applications such as mobile telephones changes substantially with changes in the acoustical load of the speaker. Presently, an acoustical solution is used for reducing the variations in the acoustical load of the loudspeaker. The acoustical solution relies on certain space and volume considerations to function satisfactorily, which are difficult to attain in today's compact mobile phones. An unconventional approach using digital signal processing to counter such degradation in frequency response is described here.
Convention Paper 6605 (Purchase now)
P15-5 Digital Signal Processing within the Steinberg VST Architecture—Roberto Osorio-Goenaga, New York University - New York, NY, USA
Steinberg Media Technologies, GmbH, of Germany, is one of the leading manufacturers of professional-audio hardware and software products. Within their software realm, they have developed a plug-in architecture for adding third-party DSP functionality for program developers who choose to support it. The architecture, commonly referred to as VST (virtual studio technology), has become a standard for third-party add-ons over the last decade, partly because of its cross-platform functionality. The software development kit (SDK) for VST plug-ins is available free of charge from Steinberg and is optimized for building within Microsoft’s Visual C++ environment on x86 PC’s, and on the CodeWarrior environment for Apple computers. This paper will focus on the implementation of classic and experimental filters within the aforementioned architecture, created and compiled on Visual C++; rebuilding these examples on a Mac should be a straightforward process. Documentation will include DSP literature as well as the process of creating VST plug-ins in a clear and understandable method. A suite of VST plug-ins will be produced and included as addendum for the project.
Convention Paper 6606 (Purchase now)
P15-6 Comparison Between Time Delay Based and Nonuniform Phase Based Equalization for Multichannel Loudspeaker-Room Responses—Sunil Bharitkar, Chris Kyriakakis, Audyssey Labs., Inc. and University of Southern California - Los Angeles, CA, USA
Traditionally, room response equalization is performed to improve sound quality at a listener. Given a loudspeaker and a listener, in a room, a loudspeaker-room response is obtained and an inverse filter is designed for loudspeaker-room magnitude response equalization. However, due to noncoincident positions of any two loudspeakers, in a multichannel setup, the combined response of the two loudspeakers may have an undesired broad spectral notch or peak or large spectral deviations in the crossover region. These spectral deviations introduced around the crossover, due to the combined phase response, generally cannot be compensated with magnitude response equalization. In this paper we compare two different methods (time delay and all-pass cascade) for correcting for the spectral deviations in the crossover region. We demonstrate that using non-uniform phase distribution, with all-pass filters, around the crossover region, as opposed to a constant phase (i.e., a fixed and optimized time delay in the satellite), it is possible to obtain better correction in the crossover region but with increased complexity. We also present an automatic approach for evaluating performance with the time-delay approach.
Convention Paper 6607 (Purchase now)
P15-7 Objective Function for Automatic Multiposition Equalization and Bass Management Filter Selection—Sunil Bharitkar, Chris Kyriakakis, Audyssey Labs., Inc. and University of Southern California - Los Angeles, CA, USA
Traditionally, multiposition room response equalization is performed to improve sound quality at multiple listeners. Furthermore, even after multiposition equalization, due to noncoincident positions of the subwoofer and the satellite, in a multichannel setup, the combined response of the two loudspeakers may include undesirable spectral deviations in the crossover region, which are different at different positions. These spectral deviations introduced around the crossover, due to the combined phase response, may be fixed by proper choice of the bass management filters. In this paper we present an objective function that can be used to characterize the performance of multiposition equalization, determine the uniformity of equalization, as well as allow automatic selection of the bass management filters for correcting the spectral deviations in the crossover region.
Convention Paper 6608 (Purchase now)
P15-8 Acoustical Monitoring Research for National Parks and Wilderness Areas—Rob Maher, B. Jerry Gregoire, Zhixin Chen, Montana State University - Bozeman, MT, USA
The natural sonic environment, or soundscape, of parks and wilderness areas is not yet fully characterized in a scientific sense. Published research in the U.S. National Park System is generally based on short-term sound level measurements or visitor response surveys associated with regulatory evaluation of noise intrusions from motorized recreational vehicles, tour aircraft, or nearby industrial activity. This paper reviews the history of soundscape studies in the National Park System and describes several recent advances that will allow automated recording and analysis of long-term audio recordings covering days, weeks, and months at a time.
Convention Paper 6609 (Purchase now)
