Subject:Pulling up vocals in field recording

Posted by: bluegreenmountain

Date:1/30/2009 2:54:25 PM

I have some recordings with a male baritone voice and medium to high pitched rattle. Sounds great with nice headphones in a quiet room, but many of my target audience will listen while highway driving. I swear that the low frequency rumble of road noise cancels out the deep-voiced singer to a significant degree. Am I making this up? And any advice on how I can edit to bring the vocalist out more? I've tried a few stabs at EQ, reverb, & compression, but I haven't cared for the results.

These are stereo field recordings, and I don't have separate tracks for the vocals & rattle, but I did have an omni condenser lav on the singer and a condenser cardioid out in front a couple of feet. Thanks.

Subject:RE: Pulling up vocals in field recording

Reply by: musicvid10

Date:1/30/2009 3:39:08 PM

You can try boosting the "presence" a little -- try +3db @ 3,000 Hz, 1.0 Octave

However, if your "rattle" is in the same range of frequencies, and I suspect it is, you are totally out of luck . . .

Another approach would be to cut the very high frequencies of the rattle a bit, and boost the overall volume accordingly.

Neither are terribly effective solutions, however.

Subject:RE: Pulling up vocals in field recording

Reply by: jumbuk

Date:2/1/2009 5:18:58 PM

This might be a really dumb question - I had assumed that the "rattle" was some kind of background noise (e.g. a cooling fan). However, your comment that the voice and rattle are not on separate tracks made me wonder - is the rattle actually something like a percussion instrument (shaker, cabasa etc) that is meant to be part of the recording?

Subject:RE: Pulling up vocals in field recording

Reply by: bluegreenmountain

Date:2/4/2009 11:22:19 AM

Sorry for not being more clear-- rattle could be good or bad. In this case it is an instrument-- a gourd rattle about 5-6" in diameter with palm seeds inside. Thanks for taking an interest.

Subject:RE: Pulling up vocals in field recording

Reply by: jumbuk

Date:2/4/2009 8:04:32 PM

Thanks for clearing this up.

I know what you mean - I have a complete set of Dylan's Theme Time radio shows on MP3 CD to listen to in the car. Dylan's voice now has no middle frequencies - it's all husky and virtually inaudible when the car noise gets up at cruising speeds.

The obvious things to try would be:

- Cut low frequencies heavily so you can turn up the volume without getting rumble intruding

- Find the mid-frequencies of the voice and boost to make them more intelligible. You would have to experiment with Q and frequency to try and isolate an area that doesn't coincide with the meat of the shaker, if that is possible.

- If the shaker rhythm is steady enough, maybe you can take a bar or two of shaker only (no voice), reverse the phase and cancel out against the full recording. You would need to accurately sync the two files though.

Subject:RE: Pulling up vocals in field recording

Reply by: musicvid10

Date:2/4/2009 8:59:46 PM

Ever try to make a recording inside a moving automobile??

You would instantly understand the phenomenon the two previous posters are referring to. It is the near-subsonic rumble, which few of us pay attention to, but which is easily measured in urban environments, that causes this effect. It is often more prevalent inside a closed automobile than standing on a busy curbside. It has been postulated, in at least one study I recall reading long ago, as being a cause of sleeplessness and irritability for urban dwellers.

**If the shaker rhythm is steady enough, maybe you can take a bar or two of shaker only (no voice), reverse the phase and cancel out against the full recording.**

That would not work. Any kind of rattle or gourd, although having a few prominent resonances, is composed mostly of non-sinusoidal waveforms, i.e., white noise. Since there is no phase for any practical considerations, there is no cancellation possible.

Subject:RE: Pulling up vocals in field recording

Reply by: jumbuk

Date:2/5/2009 3:50:36 PM

"**If the shaker rhythm is steady enough, maybe you can take a bar or two of shaker only (no voice), reverse the phase and cancel out against the full recording.**

That would not work. Any kind of rattle or gourd, although having a few prominent resonances, is composed mostly of non-sinusoidal waveforms, i.e., white noise. Since there is no phase for any practical considerations, there is no cancellation possible."

Fair argument. I guess it was always a long shot anyway.

I wonder if the Sony NR plugin would work using the shaker as a "noise sample"? The aim is just to be able to reduce the shaker levels if boosting the vocal mid-range also boosts the shaker at the same time.