Recording voice, fundamental question.

Author
NoKey
Max Output Level: -71 dBFS
  • Total Posts : 974
  • Joined: 2008/10/28 15:30:19
  • Status: offline
2009/10/12 03:50:37 (permalink)

Recording voice, fundamental question.

I compared a recording I made of my vocal and some backing. I followed most of the suggestions I have read here on clipping and so forth.

The voice is at the center.

I compared the recording to a top-notch popular comecial recording of voice and backing, and observed comparisons.

One main thing I noticed is this:
Although in the commercial recording the voice is also in the center, I seem to hear it as wider. On mine, the sound seems to be narrower, also in the center. Loudness is about the same. I like the wider sound better.

I know panning is move left or right anywhere. But what is it that I need to try to have voice sound wider, yet centered? It's hard to search for info if I don't even know the term used for that feature.

I would appreciate input on this. I believe this is an effect I want to learn about and how to apply it.

Thanks.
#1

9 Replies Related Threads

    hairyjamie
    Max Output Level: -72 dBFS
    • Total Posts : 947
    • Joined: 2008/01/23 12:14:43
    • Location: Scotland
    • Status: offline
    Re:Recording voice, fundamental question. 2009/10/12 07:41:51 (permalink)
    Try sending your Vox out to a bus with some smooth stereo reverb on it, that will widen it out a bit.

    Commercial vocals have often been double or triple tracked also to beef them up a bit.
    #2
    tyacko
    Max Output Level: -67 dBFS
    • Total Posts : 1190
    • Joined: 2007/01/06 07:20:16
    • Location: Pittsburgh, PA
    • Status: offline
    Re:Recording voice, fundamental question. 2009/10/12 08:00:12 (permalink)
    I agree with hairyjamie, it is probably just a stereo reverb panned just wide enough to thicken the vocal.

    Another neat new plug-in effect that was introduced in Sonar 8.5.1 Pro is the vocal processing effect.  They have a "Doubler" which allows you to provide an almost doubling of the vocal part.  It also has a graphical representation of how much of the doubling is occuring so you can visually see the widening as you listen.  It is an interesting new tool.

    Tom

    Our SoundClick page

    ASUS P9X79 PRO, Intel i7 3930K, 32gig RAM G.SKILL Ripjaws, RME Babyface USB, GeForce GTX 550 Ti, UAD-2, Intel 510 120gig SSD Drive, Win7 64-bit, Sonar X1E 64-bit, Studio One V2 
    #3
    Jeff Evans
    Max Output Level: -24 dBFS
    • Total Posts : 5139
    • Joined: 2009/04/13 18:20:16
    • Location: Ballarat, Australia
    • Status: offline
    Re:Recording voice, fundamental question. 2009/10/12 08:05:33 (permalink)
    It is also because it is hard to place a mono sound on a mono track well in a stereo sound field and give it the same sort of ambience that recording in a tight space might give.

    Early reflections are important and think of the dimensions of the room. If you are wanting a room with one dimension of 7 metres (20 feet) then you need to be thinking about pre delays of 20 ms and fitting the early reflections in under that time. Even early reflections under 12 ms or so will add depth to the sound. If you do it on both sides you can create quite a lively stereo ambience now from just early reflections. Reverbs are best used to create early relections. And there is nothing to stop you feeding the vocal to two reverbs. One dedicated to early reflections and the other the reverb component. eg Perfect Space. Now add in some quality convolution reverb and it can also be quite a small environment as well but mixed in low with the dry sound will give you a much wider effect now. Even tight slap delays on each side are great for giving vocals some dimension.

    The new vocal strip features doubling and it is in stereo either side of the main vox. A bit of that thrown in at low level adds a nice dimension too. See how a boring mono track is now something quite wider now and more exciting to listen to.

    When recording vocals it can be fun to set up extra stereo mikes just recording the ambience in a liveish room while tracking vocals. (good trick is to put the mike just outside the open door of a live room facing into the room. The dry sound will be pretty dry and setup a stereo pair inside the room facing away from the door) That can work a treat too behind the dry sound.

    With this much effort the vocal sound will live in its own space and be heard easily. Good engineers think about the vocal ambience if you like that is surrounding the dry sound.


    Specs i5-2500K 3.5 Ghz - 8 Gb RAM - Win 7 64 bit - ATI Radeon HD6900 Series - RME PCI HDSP9632 - Steinberg Midex 8 Midi interface - Faderport 8- Studio One V4 - iMac 2.5Ghz Core i5 - Sierra 10.12.6 - Focusrite Clarett thunderbolt interface 
     
    Poor minds talk about people, average minds talk about events, great minds talk about ideas -Eleanor Roosevelt
    #4
    The Maillard Reaction
    Max Output Level: 0 dBFS
    • Total Posts : 31918
    • Joined: 2004/07/09 20:02:20
    • Status: offline
    Re:Recording voice, fundamental question. 2009/10/12 08:11:42 (permalink)
    The new vocal plugin we got in 8.5 has a "doubling" feature that has a clever widening parameter.




    #5
    NoKey
    Max Output Level: -71 dBFS
    • Total Posts : 974
    • Joined: 2008/10/28 15:30:19
    • Status: offline
    Re:Recording voice, fundamental question. 2009/10/12 20:05:48 (permalink)
    I am very appreciative of all these replies.

    I forgot to mention my recording had reverb jazzclub effect as default from Sonar LE, so it didn't look to me as if reverb alone was the main thing. I did try vaiations of that.

    It makes sense to me about "doubling". I do have old background on waves, fourier transforms, harmonics, and advanced math for that, but am rather newbie on audio. I've been a listener most of my life. I am a rather recent home enthus on recording. I used audacity for a few years, just for simple things, no effects, just amplify, cut, paste, though.

    I am rather new in Sonar, maybe a year but have concentrated on other aspects of it, like MIDI, and MIDI maps, programing my keyboards, and so forth. That also can take a lot of time and effort.

    From readint this post, I just remember about reading somewhere that if I make a copy of a wave and shift one wave a little, the sound enhances. Is that anywere close to the term "doubling"?

    I don't have the full scale sonar and can't justify it now. I want to learn what I got so far, because it's got more than I have yet to learn to use.

    What tool can I use to shift one wave a little, if that indeed is somewhat like doubling? By how much is reasonable to shift it? I now I can shift it by hand, but that's probably not the right way.

    Another question on this: Might this have anything at all with me recording close to the mic, whereas perhaps pros record further, and are actually often (in the past) recording live musicians in large rooms as they sing?

    The recording I used to compare, by the way, is not recent at all, and I imagine those songs are recorded with live accompaniment. All those older, but good quality recordings, were analog recordings, actually (unless they have been remastered, which I've heard they do, but don't really know what they do).

    Thanks everyone, but would not feel well unless saying: special thanks to Jeff for hammering some great fundamentals in.
    #6
    mattplaysguitar
    Max Output Level: -55.5 dBFS
    • Total Posts : 1992
    • Joined: 2006/01/02 00:27:42
    • Location: Gold Coast, Australia
    • Status: offline
    Re:Recording voice, fundamental question. 2009/10/12 20:28:58 (permalink)
    Generally you want to stay away from just cloning a track and shifting, with either panned to the sides. It can sound interesting at first but has all sorts of problems with comb filtering if you try to sum it to mono, and it just doesn't sound professional anyway. So maybe try that to see what it does, but don't actually use it, cause it probably ain't gonna do much for you.

    Hairyjamie mentioned doubling and triple tracking of the vox. This is one of the big things done these days, especially in the chorus. Almost no songs you hear (at least in pop) on the radio these days will have a chorus with only one vocal take. They will often have 3 at the very least, sometimes even 10 or more. It beefs the sound up and gives the mixer stereo flexibility.

    Try this. Sing/record your main vocal. Now sing it two more times. The second and third time, try to singing it with the first vocal in your headphones, and sing it as closely to the other one as possible, especially the 's' and 'p' type sounds. Now pan the first take in the centre and at lets say 0dB. Now pan your other two takes at 50%L and 50%R. And bring their level down to -10dB or something like that. Play around with all those settings, try 100%L and R, try -15dB, try -6dB, play around with it all. Add some reverb to all of this. Maybe try a delay.

    So, the delay? Use sonitus for now, nice and easy. If you want it to have a stereo effect, don't give both left and right the same delay time. Make it different. Play around with it. Keep the delay less than 30ms to avoid noticeable echoes. for a stereo thickening purpose. Try delays of 19ms on left and 21ms on right. Try everything. Sometimes cutting the high frequency content of the delay part can result in a thicker sound that has less distinct doubling.

    On that, try cutting the high frequency content a bit of the second and third vocal take. This can make it sound less like three takes, and more like one fat take, because it is generally the high frequency content that tends to distinguish the two separate takes. Cut out all 's' and 'p' sounds from the two takes if they are not perfectly in time with the first take. Or even better, try not to sing those sounds when you do the second and third takes.

    Experiment!


    Currently recording my first album, so if you like my music, please follow me on Facebook!
    http://www.facebook.com/mattlyonsmusic

    www.mattlyonsmusic.com 

    #7
    Philip
    Max Output Level: -34.5 dBFS
    • Total Posts : 4062
    • Joined: 2007/03/21 13:09:13
    • Status: offline
    Re:Recording voice, fundamental question. 2009/10/13 01:37:14 (permalink)
    Matt,

    Great thoughts, IMHO.
    I agree with your logic for lead vox.

    I'm conditioned to use 30-32 msec delays for Hass panned backing vocs in choruses ... where things get a bit wider.

    OTOH: 19ms and 21ms may produce comb-filtering on some takes, clones, and even backing samples ... I'd be careful with all delay times under 28msecs ... depending on your resulting phasiness .

    ---------------------------------

    I Never record with 2 mics anymore ... phase issues are my greatest enemy.

    Today I routinely use a gentle amount of plate reverb ... 50msec (pre-)delays serve me very well for a thick lead vox (and heavier amounts for the snare drum)
     
    ... but done judiciously with envelopes on the reverb-send-levels.

    In sum:

    Possibly, for the lead vox: give him/her some plate reverb with generous pre-delay panned to extreme(s).

    Philip  
    (Isa 5:12 And the harp, and the viol, the tabret, and pipe, and wine, are in their feasts: but they regard not the work of the LORD)

    Raised-Again 3http://soundclick.com/share.cfm?id=12307501
    #8
    mattplaysguitar
    Max Output Level: -55.5 dBFS
    • Total Posts : 1992
    • Joined: 2006/01/02 00:27:42
    • Location: Gold Coast, Australia
    • Status: offline
    Re:Recording voice, fundamental question. 2009/10/13 01:55:34 (permalink)
    Yeah I remembered the Hass delay point where the delays become audible is ~30ms, and just kinda said some numbers under it, didn't mean those numbers specifically. I generally try to increase the delay till it is just determinable as a delay, then back it off a bit - if I'm going for that effect. But yeah, comb filtering could become a problem at ~20ms, agreed. But with a delay with a low volume it may not be very substantial. Definately something to watch out for though!

    I do like plate reverbs on the correct song. Don't really suit most of the stuff I am doing at the moment, but in the right situation, they can give you an amazing sound. Oh and early reflections, it's seriously all about early reflections. Waves Renaissance is great for dealing with them. Love it.


    Currently recording my first album, so if you like my music, please follow me on Facebook!
    http://www.facebook.com/mattlyonsmusic

    www.mattlyonsmusic.com 

    #9
    feedback50
    Max Output Level: -79 dBFS
    • Total Posts : 564
    • Joined: 2004/05/31 12:08:15
    • Location: Oregon, USA
    • Status: offline
    Re:Recording voice, fundamental question. 2009/10/15 10:20:50 (permalink)
    I generally get pretty good results with either a subtle stereo delay (very fast usually about 18ms one side and 28ish on the other). With the echo returns panned correctly it widens things up. This is in a way similar to what doubling does and both have their place.

    Doubling is more often a way to thicken part or all of a vocal for impact (or make up for a weak vocalist in some cases). It has a distinct phasing quality that varies with the lack of sync between the takes.

    I've also had good luck with specific reverb types. These are usually very subtle. They are the kind you only notice when you disable them. Combined with the right type of compression (or series of compressors) and a bit of EQ they can make a well mic'd & recorded track very compelling. Another technique (once I've got the subtle reverb working) is to add a send to an expander gate followed by a bigger reverb. By expanding the second reverb input, it kicks in mostly on the loudest phrases and has a nice way of adding a larger ambience in the places that make the most sense.

    #10
    Jump to:
    © 2024 APG vNext Commercial Version 5.1