Journal of the Audio Engineering Society

2006 June - Volume 54 Number 6


The physical measures by which acousticians evaluate the performance of rooms have evolved in large performance spaces—concert halls. They rely on assumptions that become progressively less valid as spaces get smaller and more acoustically absorptive. In listening rooms the loudspeakers and the rooms interact differently below and above a transition region around 300 Hz, similar to the Schroeder frequency in large rooms. Above this transition we need to understand our reactions to reflected sounds; below it the modal behavior of the space is the dominant factor. A review of the scientific literature reveals that natural reflections in small rooms are at levels where they are perceptible, and their subjectively judged effects range from neutral to positive. At low frequencies the long-standing problem of room resonances can be alleviated substantially through the use of multiple subwoofers, thereby providing similarly good bass to several listeners in a room. A provocative observation has to do with human adaptation to the complexities of reflective rooms, and the extent to which it allows us to localize sounds correctly in direction and distance, and to hear much of the true timbral nature of sound sources. In the case of loudspeakers, an analysis of comprehensive anechoic data is found to be sufficient to provide a good prediction of sound quality, above the low-bass frequencies, as subjectively judged in a normal room. Although the interactions of loudspeakers and listeners in small rooms are becoming clearer, there are still gaps in our understanding. A number of these are identified and are good opportunities for future research.

The ultimate performance of high-order sigma–delta ( – ) modulators is limited in view of the stability considerations that arise because the fed back decision is delayed by one sample. This means that especially as one approaches overload, the filter states can “run away” from the correcting effect of the output bit stream and the modulator will become unstable, unless a conservative noise-shaping filter design is used. These problems can be resolved if the modulator can look ahead a number of samples before making any quantization decisions. Look-ahead – modulators look forward k samples before deciding to output a 1 or a 0. The Viterbi algorithm is then used to search the trellis of the exponential number of possibilities that such a procedure generates. Alternative tree-based algorithms are described, which are simpler to implement because they do not require backtracking to determine the correct output value. Both the “tree” algorithm and the more computationally efficient “pruned-tree” and “stack” algorithms are described. In particular, implementations of the appropriate data structures for both the trial filters and the score memories are described in some detail. Although the stack algorithm offers a theoretically better average efficiency, in practice, the additional overhead required by it, combined with a relaxed requirement for a guaranteed optimal path, result in the “pruned tree” being the most efficient.

The acoustics of small rooms have been studied with emphasis on sound quality, boominess, and boxiness when the rooms are used for speech or music. Seven rooms with very different characteristics were used for the study. Subjective listening tests were made using binaural recordings of studio-produced music and anechoic speech. The test results were compared with a large number of objective acoustic parameters based on the frequencydependent reverberation times (RT) measured in the rooms. This has led to the proposal of three new acoustic parameters, which have shown high correlation with the subjective ratings. The classical bass ratio definitions showed poor correlation with all subjective ratings. The overall sound quality ratings gave different results for speech and music. For speech the preferred mean RT should be as low as possible, whereas for music a preferred range of between 0.3 and 0.5 s was found.

Effect of Reflectors on Sound-Source Localization with Two Microphones

Authors: Phatak, Sandeep A.; Ratnam, Rama; Wheeler, Bruce C.; O’brien, Jr., William D.; Feng, Albert

[Engineering Report] The localization performance of the source localization algorithms degrades in reverberant conditions. The performance of one such localization algorithm, the localization–extraction (LE) algorithm, was measured systematically as a function of the number of reflecting surfaces in a cubical enclosure. Localization was qualitatively measured using a localization plot and quantized using two objective parameters. A broad-band noise burst and a speech signal were used as stimuli. The degradation of the localization performance was monotonic but not uniform with an increase in the number of reflectors. The performance was found to be proportional to the bandwidth of the stimulus. The performance of the LE algorithm was benchmarked against that of a commonly used signal-subspace technique—multiple signal classification (MUSIC). The LE algorithm was less affected by reflections than the MUSIC algorithm. Degradation of the source localization under high reverberation was found to be more severe at low frequencies, which resulted in the detection of a “phantom” source at 0° for the speech signal.

Enhanced Multichannel Audio

Authors: Staff, AES

[Feature Article] Research in multichannel spatial audio continues to progress in novel directions. In recent times greater attention has been paid to generating soundfields in which acoustical features are more accurately rendered. This includes the representation of directional sources and a more realistic sense of depth in the listening field—aspects of which are covered in the recent convention papers reviewed here.

[Feature Article] Chairing a panel of experts from popular and classical recording fields at the AES 119th Convention, Martha de Francisco, independent producer and professor at McGill University, introduced the idea that surround sound can bring about an opportunity for enhanced emotional involvement in reproduced music. Taking an example from 14th Century art, she showed how painters moved from a lack of awareness of space and perspective (or at least an inability to represent it) to producing paintings including such factors within a relatively short space of time. With the advent of stereophonic recording, engineers had been given limited resources to represent depth in reproduced sound, but this was largely limited to representations of relative distance and fell far short of the experience of depth encountered in concert halls. Everett Porter of Polyhymnia International also pointed out that a good argument for surround is that it causes people to forget about the reproduced sound quality and start to concentrate on the music. “Surround is a good way to get people involved in recorded music,” he said. These sentiments were reinforced by Akira Fukada who thinks that the aim of the recording engineer is not only to transmit the musical performance to the listener, but also to convey the musical content in an involving way. In surround, he suggested, reproduced sound is not tiring to listen to.


29th Conference Preview, Seoul



   Registration Form

Enhanced Multichannel Audio

Surround Sound: A Chance for Enhanced Creativity


News of the Sections

New Products and Developments

Available Literature

Advertiser Internet Directory

Upcoming Meetings

Membership Information

Sections Contacts Directory

AES Conventions and Conferences


Cover & Sustaining Members List

AES Officers, Committees, Offices & Journal Staff

Advertising Insert

Institutional Subscribers: If you would like to log into the E-Library using your institutional log in information, please click HERE.

Choose your country of residence from this list:

Skip to content