2005 October, Volume 53 Number 10

CONTENTS

PAPERS

Long Interpolation of Audio Signals Using Linear Prediction in Sinusoidal Modeling
Mathieu Lagrange, Sylvain Marchand, and Jean-Bernard Rault   891
A new synthesis technique, based on linear prediction of sinusoidal parameters, is proposed as a means to interpolate missing audio in gaps lasting as long as one second. Recreated musical partials include the expected vibrato and tremolo, thereby creating a natural sound for complex polyphonic signals. This parametric approach is an improvement over the more common methods of sinusoidal and temporal interpolation. Subjects judged the result as being a higher quality.

Development and Evaluation of a Model for Predicting the Audibility of Time-Varying Sounds in the Presence of Background Sounds
Brian R. Glasberg and Brian C. J. Moore   906
A model for predicting the audibility of a target signal in the presence of a noisy background can be applied to many stimuli, such as announcement signals in aircraft or trains and ring tones of mobile telephones in urban environments. Rather than using a model that assumes steady state, this study considers time-varying signals and background sounds, which have more applications to real situations. However, the complexity of informational masking makes the problem more difficult.

Directional Resolution of Head-Related Transfer Functions Required in Binaural Synthesis
Pauli Minnaar, Jan Plogsties, and Flemming Christensen   919
When a signal recorded in an anechoic environment is convolved with a pair of head-related transfer functions (HRTFs), the listener experiences a sound source in a virtual space. An infinite number of HRTFs are needed to represent all possible locations. Interpolation among a limited number of HRTFs greatly reduces the need for a complete set, but such an approach assumes knowledge of the spatial resolution of the set and the accuracy of the resulting interpolation. In this study the spatial resolution of the basis set was determined to produce interpolated results that were subjectively identical to actual HRTFs.

New Type of Acoustic Filter Using Periodic Polymer Layers for Measuring Audio Signal Components Excited by Amplitude-Modulated High-Intensity Ultrasonic Waves
Minoru Toda   930
Because of nonlinearity in acoustic wave propagation, audio frequencies can be generated by strong ultrasonic signals. For example, two signals closely spaced in frequency create an audio signal at the difference frequency as if air were a demodulator. Measuring these demodulation products is difficult because the nonlinearity of microphones also produces similar signals. A new type of acoustic filter using periodic polymer layers greatly reduced the influence of the microphone by providing an acoustic low-pass function. In this paper a detailed theory for multilayer films as a filter confirms the approach.

STANDARDS AND INFORMATION DOCUMENTS

AES Standards Committee News   942
Tape care and handling; audio preservation and restoration; library and archive systems; audio metadata; audio connectors

FEATURES

Program Loudness Revisited   945
The Evolving World of Film Sound   951
New Officers 2005/2006   954
Review of Society's Sustaining Members   958
120th Convention, Paris, Call for Papers   992

DEPARTMENTS

News of the Sections   978
Sound Track   982
New Products and Developments   983
Available Literature   984
Upcoming Meetings   985
Membership Information   986
Advertiser Internet Directory   989
In Memoriam   991
Sections Contacts Directory   993
AES Conventions and Conferences   1000

FPO For the cover:

2005 October, Volume 53 Number 10

spine: 2005 October, Volume 53 Number 10