AES New York 2007
Signal Processing, Part 2
Poster Session P24
Monday, October 8, 1:30 pm — 3:00 pm
P24-1 Concept and Components of a Sound Field Effector Using a Loudspeaker Array—Teruki Oto, Kenwood Corporation - Tokyo, Japan; Tomoaki Tanno, Jiang Hua, Risa Tamaura, Syogo Kiryu, Musashi Institute of Technology - Tokyo, Japan; Toru Kamekawa, Tokyo National University of Fine Arts and Music - Tokyo, Japan
Most effectors used for electrical music instruments provide some temporal changes to sounds. If effectors aimed at spatial expressions had been developed, artists could have a new performance. We propose a Sound Field Effector using a loudspeaker array. Various sound fields such as a focus can be controlled in real time by sound engineering and/or artists. The Sound Field Effector is mainly divided to software parts and hardware parts. A 16-ch. system was developed as a prototype. The system can change sound fields within 1 msec. A focal pattern produced with the system was measured in an anechoic room.
Convention Paper 7307 (Purchase now)
P24-2 A Novel Mapping with Natural Transition from Linear to Logarithmic Scaling—Joerg Panzer, R&D Team - Salgen, Germany
The area hyperbolic function ArSinh has the interesting property of performing a linear mapping at arguments close to zero and a quasi-logarithmic mapping for large arguments. Further, it works also with a negative abscissa and at the zero-point. The transition from the linear to the logarithmic range is monotonic, so is the transition to the negative range. This paper demonstrates the use of the ArSinh-function in a range of application examples, such as zooming into the display of transfer-functions, sampling of curves with high density at a specific point, and a coarse resolution elsewhere. The paper also reviews the linear and logarithmic mapping and discusses the properties of the new ArSinh-mapping.
Convention Paper 7308 (Purchase now)
P24-3 Real Time Implementation of an Innovative Digital Audio Equalizer—Stefania Cecchi, Paolo Peretti, Lorenzo Palestini, Francesco Piazza, Università Politecnica Delle Marche - Ancona, Italy; Ferruccio Bettarelli, Ariano Lattanzi, Leaff Engineering - Porto Potenza Picena (MC), Italy
Fixed frequency response audio equalization has well-known problems due to algorithms computational complexity and to the filters design techniques. This paper describes the design and the real time implementation of an M-band linear phase digital audio equalizer. Beginning from multirate systems and filterbanks, an innovative uniform and nonuniform bands audio equalizer is derived. The idea of this work arises from different approaches employed in filterbanks to avoid aliasing in the case of adaptive filtering in each band. The effectiveness of the real time implementation is shown comparing it with a frequency domain equalizer. The solution presented here has several advantages in terms of low computational complexity, low delay, and uniform frequency response avoiding ripple between adjacent bands.
Convention Paper 7309 (Purchase now)
P24-4 Wideband Beamforming Method Using Two-Dimensional Digital Filter—Koji Kushida, Yasushi Shimizu, Yamaha Corporation - Japan; Kiyoshi Nishikawa, Kanazawa University - Kanazawa, Japan
This paper presents a method for designing a DSP-controlled directional array loudspeaker with constant directivity and specified sidelobe level over the wideband frequency by means of the two-dimensional (2-D) Fourier series approximation. The band of the constant directivity can be extended in the lower frequency band by using the nonphysical area in the 2-D frequency plane, where the target amplitude response of the 2-D filter is set to design the 2-D FIR filter. We discuss that the beamwidth of the array loudspeaker can be narrowed in the lower frequency band with a modification of the original algorithm by K. Nishikawa, et al.
Convention Paper 7310 (Purchase now)
P24-5 Linear Phase Mixed FIR/IIR Crossover Networks: Design and Real-Time Implementation—Lorenzo Palestini, Paolo Peretti, Stefania Cecchi, Francesco Piazza, Università Politecnica Delle Marche - Ancona, Italy; Ariano Lattanzi, Ferruccio Bettarelli, Leaff Engineering - Porto Potenza Picena (MC), Italy
Crossover networks are crucial components of audio reproduction systems and therefore they have received great attention in literature. In this paper the design and implementation of a digital crossover will be presented. A mixed FIR/IIR solution has been explored in order to exploit the respective strengths of FIR and IIR realizations, aiming at designing a low delay, low complexity, easily extendible, approximately linear phase crossover network. A software real-time implementation for the NU-Tech platform of the proposed system will be shown. Practical tests have been carried out to evaluate the performance of the proposed approach.
Convention Paper 7311 (Purchase now)
P24-6 Convolutive Blind Source Separation of Speech Signals in the Low Frequency Bands—Maria Jafari, Mark Plumbley, Queen Mary University of London - London, UK
Sub-band methods are often used to address the problem of convolutive blind speech separation, as they offer the computational advantage of approximating convolutions by multiplications. The computational load, however, often remains quite high, because separation is performed on several sub-bands. In this paper we exploit the well known fact that the high frequency content of speech signals typically conveys little information, since most of the speech power is found in frequencies up to 4 kHz, and consider separation only in frequency bands below a certain threshold. We investigate the effect of changing the threshold, and find that separation performed only in the low frequencies can lead to the recovered signals being similar in quality to those extracted from all frequencies.
Convention Paper 7312 (Purchase now)
P24-7 A Highly Directive 2-Capsule Based Microphone—Christof Faller, Illusonic LLC - Chavannes, Switzerland
While microphone technology has reached a high level of performance in terms of signal-to-noise ratio and linearity, directivity of commonly used first order microphones is limited. Higher order gradient based microphones can achieve higher directivity but suffer from signal-to-noise ratio issues. The usefulness of beamforming techniques with multiple capsules is limited due to high cost (a high number of capsules is required for high directivity) and high frequency variant directional response. A highly directive 2-capsule-based microphone is proposed, using two cardioid capsules. Time-frequency processing is applied to the corresponding two signals. A highly directive directional response is achieved that is time invariant and frequency invariant over a large frequency range.
Convention Paper 7313 (Purchase now)
Last Updated: 20070820, mei