The Effect of Ambient Noise on the Speech Reproduction of Hearing Aids
For the hearing impaired, ambient noise is a greater hindrance to the intelligibility of speech than for those of normal hearing. Fortunately, contemporary hearing aids incorporate a number of designs and algorithms to reduce the disturbing effects of noise. Still, in the fitting of hearing aids, the impact of noise must be understood so that appropriate patient counseling can take place because no hearing aid up until now has been able to manage noise problems with complete satisfaction. It is the purpose of this blog to review a few basic issues related to the reception of speech in the presence of noise, especially as it relates to indoor listening.
(Note: This blog is not for the advanced reader, but instead serves as a primer for those less familiar with this issue).
The Nature of the Noise
A common source includes other people speaking at the same time as the primary talker, traffic, machines, etc. The strength, nature, and also the direction of these noise sources determine the degree of disturbance. For those with normal hearing, direction plays a role because the organs of hearing have a selective working relationship (directional hearing) that permits a listener, as it were, to choose from simultaneous signals coming from different directions.
When a sound wave strikes a surface such as a wall, floor, or ceiling, the direction of travel is changed by reflection. These structures, especially when they have hard, smooth surfaces, can create an echo. For example, a listener sitting at the rear of a large room having these features, hears two signals: a direct signal (A) and an indirect signal (C) reflected by the rear wall (Figure 1).
Between A and C there is a delay equal to the propagation time of the sound over twice the distance between the listener and the rear wall (C). It has been established that the intelligibility of speech is not impaired if the echo is heard with a delay of less than 0.1 seconds, which implies a maximum distance of about 56 feet between the listener and the rear wall (the velocity of sound is about 1100 ft/sec. Thus, in 0.1 seconds the sound travels 112 feet, which is twice the above distance).
For a shorter distance the two signals merge so that the echo has a favorable effect in giving a higher signal strength at the location of the listener.
At greater distances the listener hears two distinct signals (echo), one after the other, and this contributes to a worsening of intelligibility.
In real life, not just a single reflection occurs, but multiple reflections are presented to the listener, meaning that the original sound is reflecting from all of the surfaces of the room again and again. As a result, the original signal acquires the character of noise instead of speech. (However, if there are only a few separate reflections, this would constitute an intermediate form in which the signal, though distorted, still has the character of speech).
A reverberation is perceived when the reflected sound wave reaches your ear in less than 0.1 second after the original sound wave. Since the original sound wave is still held in memory, so to speak, there is no time delay between the perception of the reflected sound wave and the original sound wave. The two sound waves tend to combine as one very prolonged sound wave. If you have ever sung in the shower (and I know that you have), then you have probably experienced a reverberation. The Pavarotti-like sound that you hear is the result of the reflection of the sounds you create combining with the original sounds. Because the shower walls are typically less than 56 feet away, these reflected sound waves combine with your original sound waves to create a prolonged sound – a reverberation.
Reverberation has somewhat less effect on the intelligibility of speech than an echo, but will still mask the speech if its intensity is high. Reverberation has no particular direction – it is, as it were, diffuse, meaning that it comes from all directions with about the same intensity (Figure 2).
Reverberation occurs in rooms with walls, floor, and ceiling of high reflectivity. If we produce a sharp click in such a room we hear directly after a first reflection (from the nearest wall) a number of reflections that immediately degenerate into a noise that then slowly decreases. The time that elapses between the original signal and the moment that the reverberation has dropped to a sound level 60 dB lower than the signal is called the reverberation time (RT). RT varies from fractions of a second (living room), considered to be a “dead” room, to 5 or even 20 seconds (cathedral) – considered a “live” or highly reverberant room.
Speech gives rise to a continuous noise that makes normal conversation in a reverberant room almost impossible. It is obvious that the louder the disturbing signal is compared with the direct signal, the greater the effect on the intelligibility of the speech. As the distance between the speaker and the listener becomes greater, the direct signal becomes weaker (6 dB for each doubling of the distance). However, as the reflected waves continue to ricochet between room surfaces, especially in a highly reverberant room, they lose only a fraction of power by absorption at each reflection. In other words, reverberated sound remains roughly constant because it is as if the sounds were uniformly distributed in space.
The loudness of reflected sound is less than direct sound because (1) reflected sound travels farther, and loudness diminishes with distance, and (2) reflected sound loses some energy by absorption at each reflection.
However, when the direct signal becomes weaker than the indirect sound the intelligibility drops rapidly. In rooms having strong reverberation, the greatest distance at which a conversation is possible may well be very small (perhaps 5 to 6 feet). In this situation there is no point in speaking louder because this only causes the reverberation to also becomes stronger. Figure 3 summarizes the effect of reverberation on hearing .
In general, the maximum reverberation time for clear speech is about 2 seconds. When reverberation time exceeds 2 seconds and even longer, speech becomes increasingly more difficult to understand. Speech finally becomes unintelligible at reverberation times of 4-10 seconds.
The optimum reverberation time for orchestral, choral, and church music generally ranges from 1.5 to 2 seconds; large organs 2 seconds or more, and chamber music about 1 to 1.5 seconds.
It is, of course, necessary to make a sharp distinction about noises that originate in other steady-state sources. In these cases, speaking louder does have an advantage because the ambient noise remains at the same level.
Noise from other sources may also give rise to echoes and reverberation, whereby the disturbing effect may be greater. The direction from which noise comes may again be not well-defined – it assumes a diffuse character so that the selective properties of the ear are of little help.
When counseling hearing aid patients about problems they have when listening in noisy environments, keep some of these thoughts in mind so that you can help them have realistic expectations.
- Room Acoustics, 1992, MBI Products, Elyria, OH↵