PDA

View Full Version : Distortion from using Vogon


ysh
December 30th, 2000, 05:50 AM
I have started using vogon just recently, most of the tracks that have come out were distorted in certain parts of the track, the backing vocals were turned up very high in volume (resulting in the distortion) while at the same time the rest of the music were at very low levels.

Is this a common result from using vogon? and are there any programs out there that could get rid of the distortion (crackling sound) and also a way in which the backing vocals arent so loud compared to the rest of the music?

many thanks

Brian Corr
December 30th, 2000, 07:29 AM
Vogone like all De-vocalizers is a bit hit and miss. It all depends on how the track was recorded. Some songs work perfect and some just seem to be ripped off all vocals and drums.

I am afraid its all just trial and error.

admin
December 30th, 2000, 10:24 AM
Vogone I, like all vocal eliminators (even the $1,600 one), copies the original sound and inverts the phase; i.e. what was positive voltage becomes negative voltage, and what was negative becomes positive. Then, within its processing it sums the two audio streams. Where the samples are exactly in phase and the same amplitude, the negative totally cancels the positive signal. Where they are very slightly out of phase, there can be weird effects as they don't totally cancel to zero. Some frequencies add and become louder (amplified), while others subtract and become softer (attenuated), while others totally cancel and disappear altogether.

Because you are hearing your background vocals amplified, this indicates the original recording studio engineer mixed the background vocal track slightly delayed on one channel, probably due to running it through a reverberation processor to add delay and create lower amplitude echo signals that are also delayed. If the delay is "just wrong", processing with a vocal elimator can, as indicated above, amplify or attenuate certain frequencies determined by the signal delay, which is really a phase shift in time. When the signals are inverted and added, the phase shifted signal will not exactly cancel the original signal.

On the MTU Vogone I web site pages, we have tried to be as accurate as possible in our description. Each person has a different perceiption of how Vogone will work, often based on what they "desire" it to do. There is no perfect way to remove only one sound (the main vocals) from a mixture of sounds. MTU has been working on our Vogone "II" algorithm for about a year (now you know why our current product was named Vogone "I"). We are sure it will work better, but not sure how much better. We hope to start coding in Feb-March 2001, but upgrading existing products is delaying our new developments.