I'm mastering a compilation project. The problem with this is that there are many different genres/styles and various levels of pre-mastering production quality.

There are two or three that have the lead vocal way too far out front. Yesterday I tried a new trick (for me anyway), I basically reduced the mid-band (approx 300Hz to 3KHz) which gave the perception of less vocal and more music. Then I processed the track in my normal method to get the final result.

Any comments or suggestions before I finish this project?

Thanks,

I hate to say "whatever works" but that's about all I can say... EQ is a valid way of handling it if it doesn't mess with the mix too much. Multiband compression, as much as I try to stay away from it, might come in handy also. If it (the vocal) is well centered, I'd probably split it into M/S and try working that way.

Depending, of course, on how much time you can invest in it vs. the "will anyone notice" factor.

I don't mean to sound jaded... But sometimes you bend over backwards to adjust something by a dB or two and you wonder if it was worth all thr trouble later, you know?

I actually used the multiband compressor to adjust the midband separately. The first time I did harder compression on all three bands (I typically use 3) and then pulled the mid down. I found that worked in the softer sections but in the louder sections the vocals aren't quite as clear & evident as I think they should be. Unless you told me a different trick I was going to go back and loosen the compressor on the mid-band, allowing the stronger vocal in the loud section to come through a bit more.



M/S ??? I'm not sure what you mean by this...


And a big thanks for responding!

Mid/Side (a.k.a. "Mono/Stereo")

Long story that you probably don't want to get into right now, but basically if you break up the mix into it's mono and stereo parts, you can process them individually. It's a common technique, but it takes quite a bit of experimentation and practice to get the most out of it.

But if you've got some experimentation time, try this - And forgive me / correct me if I screw up (I made a template for this and don't do it from scratch very often).

Duplicate a stereo track and line them up - sample accurate. (1&2) and (3&4).

Leave (1&2) alone. Pan (3&4) to the center (mono), invert the polarity, and reduce it's level exactly 6dB.

When you play that back together, you have only the side information.

Bounce** that to a new track (5&6)

Mute (1&2) and bounce it (mono and inverted), to a new track (7&8) Pan (7&8) to the center

Invert the polarity on (5&6) and bring (7&8) down 6dB.

(5&6) + (7&8) will give you your normal stereo signal



Now you can EQ (or limit, or compress...) the center without squishing everything else.

** If you've got the flexibility - Instead of bouncing, I send to a stereo aux at unity, then turn off the output of the channel - using the aux for playback.

I will read this again in the morning. I have just been served!

great post...

Thanks. From the context I thought it had something to do with left/right/center, like the Karaoke programs that remove the lead vocal from a CD track.

Thanks for the help!

That's the basic idea - The karoke systems and width enhancing plugs are basically multing a mono signal and twisting the polarity (the first part of the whole thing).

Yeah, the karaoke trick is nuking the middle channel.

Bear