Patrick Chevret

Work performance and mental workload in multiple talker environments

The impairment of cognitive performance resulting from the presence of speech sounds is known to increase as the intelligibility of the speech signals is improved. For that reason, speech intelligibility measures are used to quantify the nuisance potential of an unattended voice. However, most of these indexes struggle with situations in which the level of the masking sound is fluctuating. This is the case in open-plan offices, where competing voices are involved. This paper relates a set of experiments in which subjects had to carry out a basic memory task in various office-like noise settings. In addition to a target speech, the masking sounds were made up of speech and differed in level or temporal variability. Disturbance was assessed both through objective measurements of performance and subjective reports of workload. The results highlight the importance of taking into account the temporal fluctuations of the overall ambient sound when trying to ascertain the influence of speech intelligibility on observed and perceived disturbance during the performing of a mental activity. Insights are provided which could lead to the use of a speech intelligibility measure better equipped to deal with multi-sources environments.

Perception and evaluation of noise sources in open plan office

Open plan offices are now the most common form of workspaces organisation. They can improve communication between workers while saving space. Their main drawbacks are the lack of intimacy for occupants and the increase of noise level. Noise is one of the most important annoyance factor as described by workers (see SBISB study, 2010). This paper describes a study aiming at a better knowledge of most annoying noise sources in an open plan offices. It consisted in interviews and questionnaires conducted in offices together with physical measurement. This provided some information about sources and tasks for which workers are mainly disturbed. The analysis of recorded answers allowed to evaluate the influence of this annoyance on job stress and health and emphasize the influence of environmental and individual factors in the assessment of noise annoyance.

A hybrid model for echo prediction in coastal waters

Echo prediction in coastal waters is a complex task due to the fact that a very wide range of scales is involved in the propagation and the scattering phenomena. The model developed is based on a separate formulation of the propagation and the scattering aspects and a coupling procedure. The propagation modeling is performed using a high‐order parabolic operator; it is used from the source to the vicinity of the target. At this step, the field is decomposed in a set of plane waves that are scattered by the target (decomposition on the target mode). The reflected plane waves are then propagated back to the receiver. This global approach can be considered, somehow, as an ‘‘exact one,’’ nevertheless it is limited to simple shapes and is very time consuming. A simplified formulation has been developed in parallel that avoids the coupling problem by computing separately the medium and the target response. The free‐field echo of the target (monostatic response) can be either a numerical or an experimental one. ...

A numerical study of sound propagation through a turbulent atmosphere

A numerical simulation of the effect of turbulent conditions on the propagation of acoustic waves in the atmosphere has been developed. The turbulence is represented as a set of realizations of a random field generated by a limited number of random Fourier modes [Karweit et al., J. Acoust. Soc. Am. 89, 52–62 (1991)]. Through each individual realization, the acoustic pressure field is computed in the parabolic approximation. Ensemble averaging is thus performed to obtain the statistical properties of the field: mean level, standard deviation of the fluctuations, probability density functions, etc. Results will be given in particular for upward refracting conditions, when turbulence diffracts sound into the deterministic shadow zone.

A simple sound metric for evaluating sound annoyance in open-plan offices

Noise in open-plan offices has become a major health issue. Intelligible speech is considered as the most annoying noise sources by the occupants of such offices. Speech level fluctuations prevent people from achieving some high-demanding tasks, thus inducing annoyance and tiredness. Many studies were conducted in order to identify a sound metric closely related to this Irrelevant Speech Effect. Hongisto et al. have shown that Speech Transmission Effect is appropriate for evaluating the annoyance due to a neighbor in the office. More recently, Schlittmeier et al. suggested that the Fluctuation Strength can be used to evaluate the effect of the fluctuations of the ambient noise on task performance. This paper intends to present a new metric. It is based on the measurement of short-term temporal modulation of sound level. Results indicate that it seems to be as efficient as STI or FS, while being more suitable for in-situ experiments and usable by practitioners.

Noise perception of wall‐hung gas boilers

The aim of this study is to assess the quality image for domestic wall‐hung gas‐fired boilers, based on their noise. Nine boilers were recorded using an acoustic manikin in a hemi‐anechoic room, for different operating conditions. Two of these operating conditions (maximum heat input, hot water tapping) were first studied. Five‐second sequences were presented (through headphones) to sixty listeners using the mixed assessment method allowing for the comparison between signals. Analyses showed several assessment strategies: according to listeners, the relevant noise parameters could be the loudness, the sharpness or the presence of tonal components. A second experimental phase focused on tonal components by artificially modifying some sounds to offer a relevant perceptive indicator. The results of this second experiment will also be presented.

Target classification in shallow waters by time‐frequency filtering

Efficient sonar target recognition must be based on a set of relevant parameters extracted from the echoes and directly related to the geometrical and mechanical properties of the target. Previous studies have shown that a wideband approach based either on simple spectral modeling or on time‐frequency analysis is particularly suited to the understanding of echo formation mechanisms and to the classification in the free field (recognition rate as high as 96%). In shallow waters, the influence of the boundaries considerably reduces the performance of these methods due to the presence of several spurious echoes superimposed to the target echo components. Classification performance can be improved by using a method based on time‐frequency filtering matched to a free‐field reference target response. This method will be described and applied to experimental data issued from tank experiments. Several typical configurations have been first investigated: free‐field, semi‐infinite space for a set of targets (shells...