We talk a great deal about sound reinforcement for music, but speech reinforcement is by far the most common function of sound systems. However, just turning up the volume of the speaker isn’t enough. Maintaining intelligibility can be a real challenge. Here are a few things to consider.
First let’s discuss a few basic properties related to speech. Human speech is a continuous waveform with fundamental frequencies in the range of 100-400Hz – the average is about 100Hz for men and 200Hz for women. At integer multiples of the fundamental are a series of changing harmonics called “formants,” which create the various vowel sounds and transitions among them. Consonant sounds, which are impulses rather than sustained tones, occur in the range of 2kHz to about 9kHz.
Note that it’s not necessary for a sound system to cover this entire bandwidth for speech to be intelligible. Telephones, for example, utilize a bandwidth of 300Hz to 3kHz. In some very noisy environments (such as a factory), sound reinforcement systems are band-limited (equalized) to emphasize the range between 1kHz to 4kHz, where the majority of speech energy is concentrated.
There are three fundamental elements involved in satisfactory speech reinforcement:
The noise floor – Ideally, the local noise floor should be about 25dB below average speech levels for the most natural reinforcement of speech. If the ambient noise level in a space is only 15dB below the speech level, most listeners will have no trouble understanding the message, but many of them will complain about the noise level. As the speech-to-noise ratio is further reduced there will be a pronounced loss in intelligibility for all listeners, prompting sound system operators to increase the level of the reinforced speech signal.
Total SPL – There is a limit to increasing levels, however. Normal face-to-face communication is in the range of 60 to 65dB. Most speech reinforcement systems operate in the range of 70 to 75dB. If the level of amplified speech is increased beyond 85 or 90dB, there will be little increase in overall intelligibility, and many listeners will complain of excessive levels. At even higher levels intelligibility actually diminishes, as most listeners will literally feel oppressed by the too-high levels.
EQ – You can dramatically increase the intelligibility of amplified speech by making some common-sense equalization changes. First, roll off any frequencies below 80 or 100Hz to get rid of rumble. If the speaker sounds puffy, nasal, or “chesty,” try cutting a bit at 500-800Hz. Conversely, if the voice is dull or muffled, with sibilants hard to hear, a boost at 10kHz can help. Likewise, if it’s “sizzly,” with overpowering sibilants, cut at 10kHz. A thin, tinny voice can be helped with a boost at 100Hz (males) or 200Hz (females).