|1996 JANUARY/FEBRUARY, VOLUME 44 NUMBER 1/2|
Automated Parameter Optimization for Double Frequency Modulation Synthesis Using the Genetic Annealing Algorithm
B. T. G. Tan and S. M. Lim p. 3
A novel combination of algorithms has been formulated to achieve greater computational speed and improved wavetable reality. The authors draw upon Boltzmanns's energy distribution equation to simulate annealing the algorithm's genetic development, and use a genetic algorithm to select the fittest offspring. The computed results indicate the enhanced effectiveness of this approach when compared to the original sound source and other algorithms.
Gerhard M. Sessler p. 16
Tiny silicon microphones have been developed that exhibit good frequency response with low vibration sensitivity. Their diaphragm areas are in the order of 1 square millimeter. Silicon design lends itself to large-scale, low-cost production, since it can use semiconductor manufacturing techniques. It is further proposed that the silicon transducer could be integrated into an associated amplifier. Performance characteristics of several different basic transducing systems are described.
The Sound of Midrange Horns for Studio Monitors
Keith R. Holland, Frank J. Fahy, and Philip R. Newell p. 23
Extensive blind listening tests were performed in a study to identify the perceived sonic characteristics of midrange horn loudspeakers and cone radiators. The study's aim was to find the physical cause for the characteristic sound quality attributed to many midrange horn loudspeakers in studio monitors. The results indicate that the shorter horns sound more alike than the longer horns and tend to be more like the direct-radiating cone reference loudspeaker.
A Time-Frequency Auditory Model Using Wavelet Packets
Finn T. Agerkvist p. 37
Listener-perceived noise in audio coders is examined using wavelet packet analysis with rounded exponential filters and smoothed excitation through windowing. Time-frequency filters that model the human ear functions are emulated to make objective testing feasible, thereby reducing the need for extensive human listening tests to evaluate the noise-masking function. The findings indicate that the current classic masking ratio is much too optimistic. A maximum difference of 7 dB in the time-frequency pattern better represents the masked threshold for different types of maskers.
STANDARDS AND INFORMATION DOCUMENTS
AES Standards Committee News p. 51
High-density CD; preservation reports; AES-10id erratum
DRAFT AES-2id-xxxx, DRAFT AES information document for digital audio engineering - Guidelines for the use of the AES3 interface (insert)
Audio Compression for Network Transmission
Martin Dietz, Harald Popp, Karlheinz Brandenburg, and Robert Friedrich p. 58
AES News p. 82
Review of Acoustical Patents p. 54
News of the Sections p. 74
Sound Track p. 80
Upcoming Meetings p. 82
New Products and Developments p. 83
Available Literature p. 86
Membership Information p. 88
In Memoriam p. 93
Audio Engineering Society Application Instructions p. 94
AES Special Publications p. 97
Sections Contacts Directory p. 101
AES Conventions and Conferences p. 104