Real Acoustic Environment Research Articles

Sound scene data collection in real acoustical environments.

Sound scene data collection in real acoustical environments.

Audiology in an ecological perspective--development of a conceptual framework.

The conceptual framework for describing an audiological communication handicap and its underlying processes is built in analogy to a biological, ecological system and uses the following primary concepts: interaction, internal milieu, preferendum (preferred communication features) and optimal resource utilization. From these primary concepts secondary concepts and processes are formulated: signal, message/dialogue and behavior are three levels of interaction; adjustment and release of mental resources are specifications for preferendum (influenced by contextual and social conditions); specificity and facilitation underlie optimal resource utilization. The internal environment contains afferent, central and efferent speech and language processing, cognitive and emotional processes. The main components of audiology and rehabilitation (impairment, disability, handicap) are deduced from these concepts and processes. Examples of studies based on this conceptual framework are presented briefly. These include an environmental test chamber where realistic acoustic environments and communication situations can be directly studied, and a rehabilitation programme based on tutorial education for structured interaction at the behaviour level between hearing-impaired and normal-hearing persons. A vibratory-based event localization system for deaf-blind is described which improves specificity and thereby mental resource utilization.

Symmetric adaptive maximum likelihood estimation for noise cancellation and signal separation

The symmetric adaptive maximum likelihood estimation (SAMLE) algorithm is proposed. It is shown to be a generalisation of the symmetric adaptive decorrelation (SAD) algorithm for noise cancellation and signal separation. Both the SAD and SAMLE algorithms are applied to the separation of a mixture of natural speech recorded in a realistic acoustic environment. A comparative simulation confirms that the SAMLE algorithm provides superior separation capabilities.

Cochlear nucleus neural mechanisms preserve complex- sound feature encoding in the presence of background noise

Rapid sound amplitude changes—amplitude modulation (AM)—are important features of communication sounds like speech, and play a role in other perceptual phenomena including sound localization and pitch perception. Some cochlear nucleus (CN) neurons preferentially amplify AM information, relative to their auditory-nerve (AN) inputs [Frisina etal., Hear. Res. 44, 99–122 (1990)]. An important question from scientific and clinical perspectives is: How does the remarkable ability of some CN neurons to process AM change when stimuli are presented in more realistic acoustic environments containing background noise? To address this, single-unit recordings were made from 105 CN neurons and 43 AN fibers of 17 anesthetized chinchillas. Results indicate: (1) Depending on a sound’s intensity and AM frequency, a continuous wideband background noise decreased, preserved, or enhanced a unit’s ability to encode AM relative to quiet; (2) in most cases, background noise reduces the average response to AM; (3) the synchronous response to AM (fundamental frequency response) decreased in units that decreased their AM coding, and generally increased in units that preserved or enhanced encoding of AM in noise; and (4) some of these effects of noise on CN AM processing can be explained by alterations in AN AM coding at moderate sound levels, but enhancement of CN AM coding at high-sound levels appears to be mediated primarily within the CN. [Supported by NIH-NIDCD Grant No. R29 DC00408-05 and the International Center for Hearing & Speech Research—RICHS.]