6 Recording Levels
It is important to obtain the optimum level in your audio recordings. A low recording level will result in the level having to be increased either in post production or by those watching the final copy of the video. Not only will this increase the level of your soundtrack, but it will also increase the ‘noise’ level. The term noise in this instance refers to the inherent electronic noise that is unavoidably part of the signal or recording process. Normally, your intended recording will be at a much higher level than any noise, making the noise too faint to be discerned. Conversely, too high a recording level will result in distorted sound or possibly, with digital audio – no sound at all (why would this happen with digital audio?).
There are two main methods for setting the audio recording level. These are either automatic, using ‘Automatic Gain Control’ (AGC) or manual. The videographer should understand the benefits of each of these settings and in what circumstances they should and should not be used.
6.1 Automatic Gain Control
Whilst there are some circumstances in which using the AGC setting is adequate, and possibly the only means of obtaining an approximately correct sound recording, there are quite a number of reasons why its use should be avoided. The way AGC works is to constantly adjust the recording gain to try to produce a “normal” level. This is fine in situations with sound that is at a constant level, but as soon as there is a naturally quiet moment the AGC will increase the recording level. In the worst scenario, where you are trying to record virtual silence, the AGC will increase so much that it will simply record electronic noise. This undesirable characteristic is often referred to as “breathing”, and typically occurs when recording something like an interview, as it starts to lift the level of the background noise every time the interviewee pauses for thought (or to breathe).
Whilst the response times of AGC is becoming fast and increasingly intelligent, there will always be a delay in the AGC setting the correct level. If you are recording in a quiet environment and there is suddenly a loud sound, the AGC will not be able to respond to and recover from this instantly. The result will be a momentary distortion in the recording, followed by a low recording level for a moment after the sound has passed. However, if you are filming in an environment where you are unable to monitor and adjust the levels manually and need to obtain a quick, roughly correct recording level, then AGC may give you an adequate solution.
6.2 Manual Recording Level
In manual mode the recording level is controlled by the audio input gain control, which is set according to the strength of the sound source. This might require constant adjustment as the sound levels rise and fall, if their dynamic range is too high. In addition, many cameras and external audio mixers have an ‘Audio Limiter’ function which stops the recording level extending into a range which would distort the recording. In that respect a limiter is similar to the overload-reduction part of an AGC, but without the variable gain and breathing effect that AGC imparts to lower level signals. However, unlike AGC which will reduce the gain in a longer term way if high signal levels persist, limiters are only intended to deal with short-term audio peaks, and the signal can sound unpleasantly over-processed if the limiter is being over-used as a pseudo-AGC.
The main drawback of manual level control is that it requires the constant monitoring of sound levels. In an environment with constantly changing sound levels this can present a challenge.
6.3 Correct Recording Levels
Modern cameras that record digital audio normally use a meter scale with “zero dB” at the top end, representing the highest possible level before the onset of clipping distortion. This is known as a digital Full Scale meter and the values are shown as dBfs. (dBs, relative to Full Scale). The average (nominal) recording should peak to somewhere between -12dBfs and -8dBfs on the digital audio scale. Higher recording levels, or “peaky” sounds, may cause distortion as they run out of ‘headroom’ in your recording device or camera.
Headroom is a term used to describe the range between the average recording level, as indicated by the level meters, and the point at which the signal distorts. For example, if the indicated signal level is -10dBfs and audio distorts at 0dBfs, that allows for 10dB of headroom for the recording system, which is a sensible amount.
The general rule is to increase the recording level to a point at which it just reaches or peaks at -8dBfs, but take care not to let it extend beyond this point.
6.4 Additional Audio Recording Information
This section deals with additional information to help the videographer achieve the perfect soundtrack for video. As a general golden rule: try to choose recording locations that are good for sound (ie quiet and not reverberant). When you have no choice, try to adapt the environment to best suit an audio recording. Try to get machinery and other noisy background sources switched off. It’s not easy to get complete control of the working environment, but sometimes it’s a case of, “if you don’t ask, you don’t get”. Aside from the audio recorded on the camera, it often makes sense to use an external device to record a separate soundtrack. This can be used in post production to enhance the main audio recording, or to provide a wild track to ensure continuity of sound throughout the edit.
Not all cameras provide sufficient tracks for recording separate channels of audio. However, there are many devices that can be used instead, including portable MP3, hard-drive, tape and MiniDisc recorders, or solid-state devices that record onto memory cards and therefore have no moving parts. Whichever recording device is used, it can be fed from either a mixer or from separate microphones.
6.5 Practical Applications of Recording
Here are ten steps you can take to ensure the best possible soundtrack at the recording stage. Failing to observe any of these could result in a poor audio recording:
- Use the best quality microphone(s) you can afford
- Use the right type of microphone for the job in hand
- Use good quality cables
- Monitor on good quality fully enclosed headphones
- Never use VHF radio microphones unless they unless they’re “broadcast spec”; use UHF, and in both cases use diversity receivers if possible
- Adapt your environment for optimum audio recording, if you cannot relocate to somewhere more suitable
- Get your microphones as close to your subjects as possible
- Use a backup recording device if you have no room for error
- Make sure your batteries are fully charged or brand new
- Get the signal level of your recording just right. A quick rule of thumb is if you are using analogue equipment keep the signal levels higher to mask any background noise from the recorder. If you are using a digital recorder (i.e. as in all digital camcorders) which has a better signal-to-noise ratio, you can allow the signal levels to be slightly lower. In either case, you must not allow the signal to distort (‘clip’), but keep enough signal to mask the noise that’s added by the recording device.