All material © 2007. Martin Audio Ltd. Subject to change without notice.
23
The best way to overcome these problems is to have a real-life flesh and blood sound
engineer standby 24-7. This way we would always have someone monitoring the program
material who would fix every change in loudness or tone immediately. Too bad that this is
way too expensive for 99% of the venues.
What we have done to solve this issue for you is create an algorithm that reacts in the same
way as a normal sound engineer would do, hence the name ‘Engineer’. The Engineer works
with human-based perceptive algorithms which react the same way as a normal human
being would do. This means that the algorithm reacts to perceived loudness instead of
electrical loudness, so that songs with a lot of compression which sound loud will be turned
down more than more dynamic songs, so that the overall perceived level stays the same.
The Engineer algorithm features an automated volume control, an automated low EQ, an
automated high EQ and a multiband end-stop limiter to catch extreme peaks in the program
material.
Just like a normal human being would do, the algorithm listens to the incoming signal for a
couple of seconds, then decides if it sounds too loud or too weak, adjusts the volume, listens
to the amount of bass and treble in the material and, if necessary, changes it.
If the volume of the material has a sudden large boost in volume (for instance if a DJ is
fooling around), these peaks will be handled by the end-stop limiter so that the algorithm
has the time to slowly lower the volume.
The picture below is the output of the engineer algorithm with our example input. In this
picture we see that the level and tone differences have been solved in an elegant way
without destroying the dynamics or adding distortion, and that the last piece of heavily
compressed music has been lowered in volume to give it the same perceived volume.
