Sound Processing

Rarely is a sound recording made without changing the recorded track in some way. It is done both during the recording process and after the fact. Making these changes is called sound processing or sometimes sound manipulation. I prefer the former term—the latter sounds like you're doing something that's not on the up and up. In reality, the sound must be processed to compensate for many things. Inadequacies in the acoustics, a vocalist who strays too far from the mic, and the idiosyncrasies of a given microphone are just a few of the reasons for modifying the sound—and sometimes just to make the track sound more interesting.

Sound processors or effects processors come in two basic forms, software and outboard processors. Software-based processors are computer programs (often called “plug-ins” because they are typically added to or “plugged in” to the recording software) that are designed to modify the sound in some manner. As with any software, there are good plug-ins as well as bad ones. And personal preferences can enter into the decision as to which one to use. The biggest advantage of software processors is that they are digital and can be used during recording or afterward working directly on the digital signal. Anytime you convert a digital signal to analog and then back, there is some degradation in signal quality. Digital processing avoids this problem.

Outboard processors are pieces of hardware, usually mounted in the recording rack, that are used to modify the sound signal. Some are analog, some are digital, and some can be both analog and digital. A typical analog processor would be inserted in the line to modify the sound as it is being recorded. It could also be used to take a recorded digital signal, convert it to analog (called a D/A conversion), run it through the processor, and then convert it back to digital (an A/D conversion). As mentioned above, this process is not desirable because of possible signal degradation. For that reason many outboard processors today also can process digital signals to avoid the A/D-D/A conversion procedure.

So how and why do we modify the signal? Well, we'll cover a few situations here with a rather non-technical discussion. First, we'll look at compression. A compressor compresses the dynamics of a sound track. It can be thought of as an automatic volume control. Dynamics refers to the range in volume from the quietest to the loudest sounds. Don't you want to maintain a good dynamic range? Yes, but sometimes the changes in volume are due to things like a vocalist or guitarist moving back and forth from the microphone. This can be compensated for with a compressor. With proper settings, a compressor can maintain an even level of sound while preserving dynamic range. In fact, if done correctly, the compressed sound can give you the impression of having greater dynamics. With an even sound, the loudest sounds sound louder and the lowest sound quieter. Care must be exercised not to overuse compression. An overly compressed track can sound dull and monotonous.

Probably the next most important type of processing is equalization or EQ for short. Equalization is the boosting or reducing of various frequencies within a sound to achieve a desired effect. It can be used to give a vocalist more warmth or sparkle. It can be used with instruments to correct for problem sounds or to reduce extraneous sounds (such a key clicks). Equalizers come in several different types including graphic, shelving, parametric, and semi-parametric. They basically perform the same function, just using different methods. If you've ever used a graphic equalizer with your stereo to adjust the bass or treble on your system, this is basically the same effect as a recording equalizer. It must be remembered that good EQ can improve a good sound, but it will not fix a bad recording.

Next we will turn to delay. Delay is repeating the sound at a later time from the original. For example, echo is a form of delay, typically repeating the sound about a second or more after the original sound. It can occur once or have several repeats that slowly decay in volume. When the repeats occur less than about 0.7 seconds after the original, the delay is call reverberation or reverb for short. Reverb gives the sound a spaciousness without producing a distinct separate repeated sound. Reverb units can simulate the sound of anything from a very small room to a very large concert hall. A typical reverb has many settings to adjust repeats and decay and can simulate a variety of forms of reverb. The acoustics in the recording studio room may have very good reverb, in which case no additional reverb may be needed. If not, you may need to add the desired reverb. One trend today is to record vocals dry (meaning without any effects) and then add the desired reverb in post recording processing. The theory is that it is easy to add reverb, but it is difficult to remove if there is too much or the wrong kind. A track with no reverb sounds dull and unrealistic. So reverb produced from the studio accoustics or from a processor is almost always required.

Related to equalization is filtering. Filtering removes only a certain band of frequencies. For example, you can remove high frequencies with low pass filtering (LPF) or remove low frequencies with high pass filtering (HPF). You can even remove a band in the middle (notch filtering) or remove all frequencies except for a specific band of frequencies—band pass filtering (BPF). Filtering is used to remove or reduce hiss, buzz, clicks, pops, and other noises that can slip into recorded sound. For example, many microphones have a switch for high pass filtering to remove rumble that can occur from vibrations in the studio. Some preamps and other processors have various switches for removing various offending sounds.

Other effects include noise gates. Noise gates shut off the audio signal when the sound level drops below a certain preset level. This process reduces noise from the recording during quiet passages. Another effect is chorusing, which makes one sound be perceived as two or more similar sounds. Detuning is similar to chorusing, but it adds a slightly detuned signal to the track resulting in a fatter or thicker sound. Phase shifting is an electronic effect in which various bands of frequencies are shifted in time relative to each other. The sound produced has a motion or a swooshing effect. Pitch shifting can be used to correct an out-of-tune note. Harmonizing can produce several outputs, each with a different pitch that is in harmony with the original pitch. These are just a sampling of the many effects that are available.

When used properly, effects processors can be used to modify, improve, correct, and enhance a recording. When used improperly, processors can make a recording sound awful. The difference is in the skill of the operator/engineer. If you have questions about sound, please contact us.

[Articles] |  [Home]