Piano processing

[shure_1302]

Hi!

I am recording piano and vocal. Piano is recorded from the stereo output from my Roland. Is it any way to make piano sound better? In otehr words is there any standard piano processing in general (compression, equalisation , etc)? I understand that there is now general solution for that but may be you could refer me to any web info.

And more general question - if I want to apply reverb for both vocal and piano should I use the same type but may be at different levels?

Thank you indeed.


Roland HP107e - Presonus Firebox - Cubase LE4

[brandondrury]

Is it any way to make piano sound better?

Not exactly. When you take the line out of a unit like your Roland that IS what the is. (I have a big section in my upcoming home recording book about "It Is What It Is".)

You can EQ it, compress it, or distort it, but there is no "better knob" unfortunately. I had a band yesterday griping about the piano sound of the Kurtweil the kid used. He wanted it to sound "better". Again, it is what it is. The best way to make it sound "better" is to use MIDI to trigger killer piano samples.

In otehr words is there any standard piano processing in general (compression, equalisation , etc)?

Maybe on a real piano, but the line out on a synth is supposed to be ready to go.

if I want to apply reverb for both vocal and piano should I use the same type but may be at different levels?

There aren't any rules for this one. It depends on what you want. If you want them to sound like they are in the same space, you definitely want to use just one reverb as an aux send and then vary the amount of signal you are sending from each track. In Cubase LE4, this is called an "FX Send".

Brandon

[shure_1302]

thank you very much.

Yes you are right there is no "better knob". In fact I am more or less satisfied with the quality of the sound per se - I guess it is very close to the real piano. I am more concerned about the "processing" for vocal+piano songs. I mean because of the different playing expression piano sounds too loud or too quiet in the same song - should I compress it or just leave it as it is? Or use the volume automation?

Also what is the standard trick to make voice more upfront and piano (or any other accompaniment) more on the back (or sides?)? When people say "pan off" what does it actually mean?

thanks.

[markiegiscool]

"Pan off" is sending the sound either more to the left or right of the stereo mix. 99% of the time you would have your vocals up front, and pan the instruments. However, with just a piano and vocal I would be reluctant to send the piano too far one way or the mix might seem empty.

Are you taking a stereo signal from the piano? With just a piano/vocal mix I would send each channel of the piano hard right and hard left and keep the vocal in the centre.

More reverb on the piano and less on the vocal will make the piano seem more in the background, and the vocal upfront.

With a song that has just the one instrument and vocals, the different playing expressions should work well. Rather than adjusting the piano's volume, can't you sing more to match it? I love the impact of a piano when it changes from a quiet to a loud moment. I imagine if you compress the piano too hard you will lose all of the feeling and delicate-ness from the playing.

[shure_1302]

Thank you indeed!

So "pan off" means to send teh signal to the one side?

I am taking two jacks outputs from the piano and record it as single stereo track in Cubase. So I guess it is already in the state you have described.

With quiet/loud issue - yes exactly, when the piano is loud the voice is loud as well. But then the song become too dinamic - so may be the compresion of the mixdown is good? What kind of compression should it be?

thanks.

[markiegiscool]

Yeah, as far as I know the stereo track in Cubase is just panned hard left and right. I never use them...Not really sure why. I always record to two separate mono tracks.

I'm no expert on the whole compression thing, you maybe ought to wait a bit until a few of the Americans on here wake up, and they might be able to advise you!

[Ruzz]

Originally Posted by shure_1302

I am more concerned about the "processing" for vocal+piano songs. I mean because of the different playing expression piano sounds too loud or too quiet in the same song - should I compress it or just leave it as it is? Or use the volume automation?

Depends. Decide whether the playing is on the whole "good" or erring toward "not so good" - both relative terms but you should have a good idea of whether the results you hear are a result of bad playing or not. (And of course, if the playing is truly bad - no quotes - a retake is in order). Where you draw the line between good/no-so-good playing is a judgement call - again, down to you.

If you think it's more a playing problem (erring toward the not-so-good), I'd try compression first - using settings to squash the more extravagant parts. But be careful not to remove the "human" qualities - you want enough compression, not too much. Secondly, by all means, try volume automation - but if you're doing a lot of lowering and raising, by significant amounts... perhaps a retake is in order?

Another thing you could try is... double up the piano track in the DAW. Compress the shit out of one of them (ok, maybe not that much ) and the other bring up to a level where the louder parts just "poke through" in the mix. This certainly works well for some things - never tried it on piano.

However, before you try any of that...

Also what is the standard trick to make voice more upfront and piano (or any other accompaniment) more on the back (or sides?)?

Well, in one respect, that's what reverb typically does - more reverb = "further back". The more you add tends to do two things (IMO) - make the source (instrument/voice) sound "smaller and further away" while the space it's in sounds "bigger".

You asked earlier about using the same reverb for both voice and piano. Here's my take on it.

Piano has two sets of reverberation - it's own intrinsic reverb (coming from the big old coffin itself) and the reverb from the acoustic space in which it sits. The question I'd ask myself is - does the intrinsic reverb sound like a real piano "up close" ? What I'm suggesting is using two reverbs - one for the overall space - which piano and vocal share - and perhaps another reverb just for the piano. This reverb is deliberately "boxier" sounding, tighter... feed this sound (piano+boxy reverb) into the shared one. Whether this will work with your Roland I can't say - depends what you're getting from it "raw" - if the signal you get from it already has an obvious "ringing coffin" sound then ignore what I just said - adding another will likely make it worse.

Now, if you get a great reverb sound, it may just be that you don't need to do too much compression/volume automation. I'm suggesting that you first record another piano track (just for test purposes) and play with reverb combinations. If you find one that works, save a preset and then apply it to the problem piano. You never know, that expression you thought was heavy handed/too light in places, may just "melt away" with the space correction provided by reverb.

And then again, maybe not

Good luck!

Ruzz

[shure_1302]

The playing is o'k, I guess. The question I meant is simply "What to do with very dinamic song?". In some parts the piano _should_ be "forte" and in some should be "piano". However when it is in combination with vocal the mixdown sounds strange - dynamic diapason is too high.

However the second highly compressed track may be a solution. Will I get the same result if I simply put high compressor in Send insteaf of doing two track?

I shall attach the mp3 with the piano - I guess they already included "coffin reverb" effect in roland piano samples.

Thank you for the information.

[Ruzz]

Originally Posted by shure_1302

However the second highly compressed track may be a solution. Will I get the same result if I simply put high compressor in Send insteaf of doing two track?

Well, there's nothing wrong with that as an approach - the intention is the same, after all. So, try it! ... they are different though, so what the hell, try both!

I guess they already included "coffin reverb" effect in roland piano samples.

Surely

Thank you for the information.

Information? I'm just making it up as I go along

Ruzz

[richiebee]

If a composer writes for the piano to get loud, they would normally get the singer to get loud too. Otherwise, what's going to happen? The singer will get drowned out. So, is the problem a singer who isn't following the dynamics? A piano player who isn't sensitive to the dynamic range of the singer? Or a composer who forgot to put dynamics in the vocal part?

If its just piano and voice, compression on the piano should be very light. Because of the attacking nature of the piano, the effects of compression are heard very quickly. I would think that if this is a good performance, you should compress the voice instead, then bring up the whole vocal track. You can then use faders to make any further adjustments. The overall track might need some compressing to have the whole thing at a listenable level, so just go easy on the piano only compression...

Using compression as a send effect is not the same as using it as an insert effect. You might get an interesting result from using a compressor on a send, but it certainly won't be a natural sound.

So, what is the genre anyway?