How do you "de-mush" 5 part harmony vocal

[bluessinger45]

Hello all,
I'm recording a song in Audacity and wondered how I could add more "space" to the harmonies (if this makes sense)!
I'm harmonising with myself ,mainly different variants of 3rd and 5ths and it all starts to mush a little.
I have applied a small amount of reverb on the way in via Edirol UA4FX to each track.
Other than that I have done nothing yet in terms of eq or compression.
Style is folky wolky (my description) with at the moment 3 acoustic gtr tracks along with the vocals.
any ideas gratefully accepted

[jollyon]

My answer would be by adding some amount of contrast to each of the parts / groups. Lots of different ways to achieve this. Height (frequency content), width (panning), depth (reverb / delays), tone (tapes, tubes, valves) and timing (attack / release times). If it's for BVS that are meant to work together as a whole, it might be a small amount of panning, tone and / or very slightly different attack / release times on a compressor.

A few links below which I think are well worth checking out:

JJP on compression for musicality & timing (@30 mins). He also covers his mid range philosophy (@23 mins)
- http://www.veoh.com/watch/v65188118rSrRSN8Q

Great article on controlled use of reverb
- http://ledgernote.com/columns/mixing-ma ... th-reverb/

Ganesh Singram covering his mix zones philosophy
- https://www.youtube.com/watch?v=9ayteRHOMBQ

Lastly, next time you're recording you could try different mics / mic placements / recording techniques to accentuate difference before mixdown.

Hope this is helpful and good luck with your project.

[bluessinger45]

Thanks systemheavy for those links - I'm taking a good look at them now. Hopefully will produce some improvements and more clarity

[Drew Stephenson]

Is it 5 equal parts or a lead part with backing harmonies? If it's the latter then you can actually be fairly hefty with your eq, even in a fairly sparse mix, to make sure that the harmony parts are really only contributing in the bit of the spectrum you want them in.
Adding a bit of pre-delay on the reverb as well can help make the lyrics clearer as well without affecting the overall feel.

[Freuman]

Another tip is actually in the recording process; rather than close micing the harmonies, take a few steps back.

When you have multiple singers (like a barbershop) being recorded at once they tend not to all stand in the same place. Adding real space will also add space in the mix.

[jollyon]

Is it 5 equal parts or a lead part with backing harmonies? If it's the latter then you can actually be fairly hefty with your eq, even in a fairly sparse mix, to make sure that the harmony parts are really only contributing in the bit of the spectrum you want them in.
Adding a bit of pre-delay on the reverb as well can help make the lyrics clearer as well without affecting the overall feel.

Hey blinddrew. I'm with you on that. Even a few milliseconds of pre delay on a verb can avoid smearing the front of the transients. there's some helpful info on this in the reverb article.

[BJG145]

+1 for the suggestions above. You might want to experiment with different mics/placement if you have any available; bear in mind that any problems with the recording, eg room modes, will be multiplied when stacking up similar takes. Harmony vocals generally need careful tuning and alignment. You can be generous with the de-essing, and thin out or delete breaths between phrases.

[bluessinger45]

Guys - thanks for these responses - very helpful.
In the style of Dennis Waterman I have played all the instruments , sung all the vocals and written the tune.
I guess this means that I can "place" my own bv's using distance from mic in future.

I am confident I can get to a good result as I am using Audacity as a recording tool rather than a Digital tinkering tool.

My view is that Fairport Convention produced good results with 4 or 8 tracks and little other equipment so I should be able to do the same !

[Rattleshock]

Without hearing the 'mush' you are referring to, it's hard to comment directly. For me, generally speaking, I process all BGV's (when used behind a lead vocal) together. I will get the panning sorted out on the individual tracks first, which I typically like very wide, and then bus them all to a stereo pair. I like to compress/EQ because the stereo pair because I'm not usually looking for voice distinction within the BGV's and this method allows me to treat them as a single instrument.

Some things I regularly do:

- Low cut the BGV group until the muddiness is gone
- A broad high shelf boost until I have the crispiness and cut I'm looking for. If it becomes too bright, I will then high cut back into the boost, which leaves the broad lift but can take out the harsh and brittle ultra high frequencies.
- Depending on the recording and the lyrics, a de-esser would be next
- Compressor (if needed)
- Any time/space processing (delays/reverbs/etc)

Time alignment of BGV's is very important, especially if they are working together to support the lead vocal. It's not uncommon for me to edit starts and stops of words so they work better in unison. I also usually remove harsh consonants from all but one or 2 BGV tracks (letter 's', 't', etc.). Sometimes I'll even cut them right out of all the BGV tracks. The last thing you want is to start stacking "S"s and hearing them ping out of time across the stereo spectrum. I also am pretty liberal with removing breaths, and leave that for the lead vocal only. It all depends on the song and the style. But this is my usual approach. Hope it helps you.

Cheers!
Glenn

[bluessinger45]

Without hearing the 'mush' you are referring to, it's hard to comment directly. For me, generally speaking, I process all BGV's (when used behind a lead vocal) together. I will get the panning sorted out on the individual tracks first, which I typically like very wide, and then bus them all to a stereo pair. I like to compress/EQ because the stereo pair because I'm not usually looking for voice distinction within the BGV's and this method allows me to treat them as a single instrument.

Some things I regularly do:

- Low cut the BGV group until the muddiness is gone
- A broad high shelf boost until I have the crispiness and cut I'm looking for. If it becomes too bright, I will then high cut back into the boost, which leaves the broad lift but can take out the harsh and brittle ultra high frequencies.
- Depending on the recording and the lyrics, a de-esser would be next
- Compressor (if needed)
- Any time/space processing (delays/reverbs/etc)

Time alignment of BGV's is very important, especially if they are working together to support the lead vocal. It's not uncommon for me to edit starts and stops of words so they work better in unison. I also usually remove harsh consonants from all but one or 2 BGV tracks (letter 's', 't', etc.). Sometimes I'll even cut them right out of all the BGV tracks. The last thing you want is to start stacking "S"s and hearing them ping out of time across the stereo spectrum. I also am pretty liberal with removing breaths, and leave that for the lead vocal only. It all depends on the song and the style. But this is my usual approach. Hope it helps you.

Cheers!
Glenn

Thanks for that Glenn. I inadvertently followed a lot of your recommendations by accident the other night as I spent a while sub-mixing the chorus to its own mix, though I did leave the Lead vocal in . I also chopped about so all phrase endings were at the same time , so a good tip there as well.
I'm starting to think I will dare to post the final version for "constructive criticism"....

[pariah223]

Make sure your takes are as tight as you can possibly make them. You will be amazed at how easy vocal harmonies are to mix if they are solid and tight. But aside from that. Panning helps as well. If you double some of the harmonies and pan them (as opposed to a stereo seprator which can work with less predictable results) and think of reverb as a foward to back panning knob to create more space. Also, composition is key.. 5 part harmonies are great but van be even more effective if those harmonies only stress certain words and the rest is 2 or 3 part harmonies. Depends on the style tho. Ill comment more when i hear your song. Look foward to it

[bluessinger45]

Here's my final mix of the song - I made a video of accompany the music.

https://youtu.be/sdF587VS0hw

[sixtysixmix]

Here's my final mix of the song - I made a video of accompany the music.

https://youtu.be/sdF587VS0hw

I had a quick listen and I didn't hear any mush

One thing I did notice, however, is that the harmonies are all singing the same exact timing as the lead most of the time. I think they would be more pleasing to the ear if there were some differences where the BV's drag out a note or something now and again? Or even sing to a different timing or a sub-set of the words. Not sure what you were going for.