MP3 Recording of Live Performance - Need a Little Help!

Q

quesne

Member
Hello All:

The attached file is a stereo Zoom H4 recording of one song from a little concert my partner and I did a few days ago. The problem (as I see it) is that the guitar and vocals are quite muddy and far behind in the mix, while the mandolin is better defined and pretty much in front of everything else. This probably has mostly to do with mic placement. I used the built-in H4 mic array, at a 90 degree angle. The recorder was placed about 5 feet away (toward the audience), three feet off the ground and well centered in between us.

I am so accustomed to recording in my home studio with each track independently managed, I am not very adept at dealing with a "fixed" mix. I was hoping that someone could help me get the vocal and guitar a bit more forward in the mix and better clarified. If anyone has a moment to listen and give me some advice, I would greatly appreciate it!

Thank you in advance :-)

P.S. The only post-processing I did in Reaper was to normalize the waveform, and remove noise via the "subtract" mode in the ReaFir plug-in. If the unadulterated audio is needed, I can of course provide it. Thx!
 

Attachments

Last edited:
Well, the mic probably picked up how it sounded. A few years ago that would have been that, but now there is some potential to use software to unmix the recording.

Hopefully you recorded it as a 24 bit wave file. Mp3 is not good for music production. It was developed for music delivery in a world with dialup internet.
 
Understood. And unfortunately, that’s very much the issue. I wish I had thought to set the file type to WAV before hitting record. But I didn’t. Hopefully the last time I make that mistake.
 
I just keep my H5 on 48/24 wave. It has literally never been set to record mp3 format.

You've got two additional inputs. Next time, perhaps you could have supplemental guitar and vocal mics. Or just place the mic to compensate for the loud mandolin.
 
The real thing to consider is what it’s for. Two instruments and two voices, or two instruments and one voice plus the audience? One mic per source, mix afterwards. Good separation, panning , scope for eq and effects, clarity and definition. The alternative is a stereo recording. That requires balance. It sounded like that where the mics were. We cant tell if that would have been different sitting somewhere else. Its quite possible your voice was simply not loud enough, anywhere. I doubt you deliberately put the recorder near just one instrument? The essential thing with capturing acoustic live playing is to get the balance right. Thats pretty well impossible for anyone who is playing, without endless move, record, replay on phones, and repeat cycles. You need a person with closed back headphones who can find the magic place, if there even is one. Live, our brains can filter out an over loud source. Recorders can’t. I ran it through spectral layers which is good for unmixing, but it could not separate guitar and mandolin at all. You perhaps should consider a small PA, after all, your recording shows what the audience got doesn't it?

They are struggling with picking out the vocals in the same way we are? With a 4 channel recorder, I’d record all the sources and have an audience mic for the ambience and applause. My first live recording was exactly like this. A four piece trad jazz band. Virtually the same sound as yours. That was in 1976. I knew all the science and had the kit and recorded in stereo, expecting success. It was a shock what I recorded didn't sound like my ears told me. Since then, I cheat, recreating the live sound back in the studio. Stereo recording for me is choirs, small orchestras or ensembles, in nice spaces. A solo opera singer, or guitars, or even a recorder consort needs a really good listen in the space, and if they want a really good, commercial sounding recording I will probably cheat. In many of these situations there will be one player that does not project and always one who’s instrument cuts through like a siren. There will be one singer with a wonderful but quiet voice and another with an instantly horrible nasal tone, but who is giving it 110%. Much as I hate it, i would prefer an excellent cheat than a mediocre real capture in revealing, but unrepairable stereo.
 
I noticed that the mandolin is very much strong and centered, so on a whim, I split the track and inverted one side. It's an old "remove the vocal" trick that was supposed to take out a vocal that is centered, and leave the stereo instruments.

I then played with panning and levels to see what I could get. I had to EQ it and compressed it to fatten the sound. I don't know if this sounds better to you, but it might be a bit more balanced.
 

Attachments

Last edited:
You must log in or register to reply here.
Back
Top