One way to have proper loudness is to measure the audio and use this measurement to set the dialog loudness metadata parameter--and do this for every program that is broadcast, which may or may not be practical or possible. Another way to accomplish the same goal of controlling loudness is to adjust the audio signals to match a single, static dialog loudness parameter.
Sometimes known as audio processing, this technique involves adjusting the average loudness of a program on a sample-by-sample basis, and may also involve reducing the dynamic range of the audio program. In other words, programming that is quiet is increased in gain while programming that is loud is reduced in gain. Done properly, this will create a consistent program that can closely match a single, non-varying long term, A-weighted loudness equivalent value (LeqA).
Audio processing is a remarkably effective technique for controlling loudness, but it does irrevocably change the original audio programming. This may or may not be a desirable side effect. If the goal is to control loudness and also maintain the original audio, the solution is more complex, but it is possible.
MOST READ
