| 2007 October, Volume 55 Number 10 |
CONTENTS
PAPERS
Implementing Asymmetrical Crossovers
Neville Thiele 819
Crossover networks for loudspeakers are often designed with symmetrical pairs of transfer functions, but there are applications where the low-pass channel can use filtering of lower order than the high-pass channel, remembering especially that the tweeter's transfer function should be incorporated into the criteria for the crossover design. Two kinds of networks are described, and the advantages and disadvantages of asymmetrical crossovers are discussed.
Ultrahigh-Resolution Audio Formats for Mastering and Archival Applications
M. O. J. Hawksford 833
With the increasing number of audio data formats available for final release, efficient archival storage of the mastering source material becomes a problem. Ideally, regardless of the initial format, archival storage should not degrade the dynamic range or spectral bandwidth. Several high-resolution strategies are presented as candidates for high-quality archival storage. One candidate, multilevel SDM, achieves the desired goal. An intermediate representation, which can be converted back into any of the standard formats, takes advantage of the lack of energy at very high frequencies.
The Effect of Stereo Crosstalk on Intelligibility: Comparison of a Phantom Stereo Image and a Central Loudspeaker Source
Ben Shirley, Paul Kendrick, and Claire Churchill 852
Intelligibility of speech is improved when presented through a real center loudspeaker rather than with a phantom center image from a stereo pair. A two-channel format is less intelligible because the acoustic crosstalk effect produces a dip in the frequency response for mixes that pan the speech to the center. Subjective tests showed a statistically significant increase in word recognition accuracy when presented through a real center loudspeaker. Not only does a center loudspeaker produce better clarity, but it also improves intelligibility.
Ambisonic Synthesis of Complex Sources
Dylan Menzies and Marwan Al-Akaidi 864
A method for encoding a general acoustic source and transcoding it for a high-order Ambisonic format is presented. Although the approach is more elaborate and costly than plane wave or monopole synthesis, the effort is worthwhile for complex sources in quality rendering systems. For example, localizable sources differ from simple monopoles because of nonzero width and nonuniform directivity patterns. The proposed approach can also represent reverberation.
STANDARDS AND INFORMATION DOCUMENTS
AES Standards Committee News 877
Audio metadata; loudspeaker modeling and measurement
FEATURES
31st Conference Report, London 882
Intelligent Audio Environments 889
New Officers 2007/2008 896
Review of Society's Sustaining Members 900
124th Convention, Amsterdam, Call for Papers 927
DEPARTMENTS
Reviews of Acoustical Patents 879
Upcoming Meetings 899
News of the Sections 916
New Products and Developments 919
Membership Information 920
Advertiser Internet Directory 921
In Memoriam 926
Sections Contacts Directory 928
AES Conventions and Conferences 936
FPO For the cover:
2007 October, Volume 55 Number 10
spine: 2007 October, Volume 55 Number 10