Acoustic perception and sound quality

For a long time, the acoustic engineer's problem was to reduce the noise emitted by the object under development, on the assumption that this reduction in noise level would lead to a reduction in annoyance. This principle is, of course, often verified; but, in some cases, it may be more effective to work only on certain aspects of the noise timbre, thus improving the ratio between the cost of the acoustic treatment and its perceptive effectiveness. Moreover, the noise emitted by the object can also be a source of information for the user: in a car, for example, the slamming of the door must be loud enough for the passenger to know that the door is closed.

This is the notion of acoustic quality, which represents the balance between the sound evoked by an object and the general image its designers want to give it.

To improve acoustic quality, perceptual analysis approaches can be implemented, providing a useful complement to conventional vibration and acoustic characterization and optimization methods.

These perceptual studies have the following objectives:

• firstly, to identify the aspects of timbre that are conducive to the acoustic quality of the product (or those that are the cause of discomfort felt by listeners). This knowledge helps the product designer, for example, to quickly select the best ways to improve the product;

• secondly, to develop reliable indicators of noise perception, which can be used to define specifications for new products.

The difficulty lies in the fact that the relationship between sound, a physical quantity as conventionally measured (or calculated), and its appreciation by a listener is often complex. The process can be illustrated as shown in figure 1 .

First of all, as will be explained later, the sound picked up by the listener's eardrums is different from that measured by a microphone. This signal is then interpreted by the listener, who attempts to recognize the source that emitted it; this recognition phase is essential, as a sound cannot be considered independently of the phenomenon that gave rise to it.

The third phase involves the listener determining, even implicitly, a few salient characteristics of the sound (which may appear loud, regular, high-pitched, etc.).

This knowledge enables the listener to assess the noise, a phase which is eminently subjective, since the influences of context and the listener's previous experience are very important. In the field of environmental noise, we know, for example, that some people...