How would you handle this difficult situation?

4

4tracker

Guest
I'm working on several projects, one is an old recording bandmates did straight into a boombox with cassette. We were a zero budget high school punk band. So it's an awful recording, with guitar, vocal, and sometimes bass all playing at once into a boombox mic. There is a ton of low end from the bass, and then the opposite problem of shrill/harsh guitar on the upper end. Obviously EQ (cut the low, and notch the 5k region) helps a lot, but is there any other trick? Also, the vocals are very low on several tracks.

I'm new to mixing and teaching myself as I go.

I'm reading about parallel compression. Would this help in my situation? Mainly, would it boost the low vocal, which I could then mix into the original track to regain some transients? Is there any trick or tip you guys have to help with poorly recorded material like this?

To recap, the three problems are:

1. Low vocals
2. Loud bass
3. Shrill guitars

What is difficult is all are on 1 track.

2 and 3 can mostly be resolved with EQ.
 
4tracker said:
..I'm reading about parallel compression. Would this help in my situation? Mainly, would it boost the low vocal, which I could then mix into the original track to regain some transients? Is there any trick or tip you guys have to help with poorly recorded material like this?

To recap, the three problems are:

1. Low vocals
2. Loud bass
3. Shrill guitars

What is difficult is all are on 1 track.

2 and 3 can mostly be resolved with EQ.
Click to expand...
The basic is that any compression (straight or 'parallel) brings up all lower level stuff -relative to what you had. How do you get it to target mostly just the vocal? Maybe, if you can get lucky with finding it in a specific freq band?
 
Without hearing it (which would be really helpful)...

My first thought for the low end would be to, hopefully, find a frequency or two that is a loud spike. So I'm guessing there is some frequency between 100hz and 180hz that you could cut. I'd make it very deep, like 6-8dbs and very narrow.

The bigger problem is going to be the combination of shrill guitars and soft vocal. They're probably sharing a lot of frequency space, and so what you do to one is going to affect the other. I'd probably try to notch out something around 2400hz to tame the guitar. Then I might try a broader, but small boost centered around 6000hz to help bring out some of the vocal.

Don't take these numbers as failsafe advice. This is just kind of a general approach I'd take and the general areas I'd look at.

As to your specific question about parallel compression. The only thing I would even consider is using it to try to bring out the vocal. So maybe:

1. duplicate the track.
2. roll off the lows on the dup, like all the way up to 3Khz.
3. Compress the snot out of what is left
4. Try to blend that in with the base track.
5. Don't try to convince yourself that this technique improved things just because you read a cool article about it.

Or just replace the recordings. :) WTF?
 
TripleM said:
Or just replace the recordings. :) WTF?
Click to expand...

Lol. Yeah we were 16 and we're almost 40 now. And don't live near one another. No way we duplicate that energy/teen angst.

I basically did what you said. It sounds much better, but the vocals are the issue, since as noted they share much of the same frequency as guitar. I was wondering if there was some secret I was overlooking...

I'll try the compression idea, but I don't have much hope that will sound good.

I'll try to upload the files later today.
 
Last edited:
Cool. Put it up in the MP3 clinic. See what people there say.

Or just replace the recordings. (that joke gets better every time someone tells it).
 
well firstly this is mastering not mixing, and secondly you can't change the balance of a mix later. I would probably use M/S EQ for this kind of thing, if you don't know what mid/side is, it's worth searching the interwebz
 
Can the admin move this to the mastering sub forum?

The mid-sid eq looks cool. I never even heard of it. Is that commonly used, or only for difficult jobs?
 
Neh. Just because it's 'two track doesn't make it 'mastering'. More forensic audio.. or mix/audio problem solving.
Besides it's all the same tool boxes :)
 
mixsit said:
Neh. Just because it's 'two track doesn't make it 'mastering'. More forensic audio.. or mix/audio problem solving.
Besides it's all the same tool boxes :)
Click to expand...

It really does feel like forensics.

I tried the mid-side eq tonight. I need time to master the knobs and figure it all out, so I'm not sure if it helped yet. I just didn't have a lot of time tonight, but I'll try that more tomorrow. My other idea was to duplicate the tracks and compress one severely, hopefully bringing up the vocal, then mixing it with the original for the dynamics. I have no clue if that'll work. I realize I won't get a technically good sound no matter what, so all I am looking for is improvement, and mostly just a more audible vocal.
 
Your biggest problem will be the compression the boom box used for auto-levelling. The compression will be riding the low end, so simply eating out the lows will leave something thin and pumpy.
 
VomitHatSteve said:
Do you have the tracks to post so that others people can take a stab at it?
Click to expand...

I do, and I can do that later today. They are immature, super lo-fi punk/folk songs written by 16 year old versions of us. haha. I just figured people would criticize the tunes because the internet can be crazy. But sure, later tonight.
 
M/S processing won't be of any use unless there's substantial stereo separation, which is unlikely with a boombox recording.
 
As stated, depending on the method of recording and bussing, M/S may do nothing.

One thing that may help here is multiband compression. Obviously not ideal, but given the situation, it may help.
At least you could target specific ranges, cut/boost/compress/expand,range them differently to effect the fundamental frequencies of some parts separately..... again, obviously not ideal, and obviously limiting (no pun intended).... but may help some

I use Waves C4 or C6 for this a lot, as I can even specify RANGE, and also expand, all in one plug.... I get files brought in that can't always be re-recorded (usually budget concerns, or concerns like yours)

Another thing that sometimes helps is to duplicate the track a few times (varies), and set freq ranges with high/low passes, so each track is it's own range - sort of multibanding really, but on separate tracks so each range can be effected on it's own with other plugins as well... this can be tedious, but sometimes helpful
 
You must log in or register to reply here.
Back
Top