How to widen vocals space/perception?

[Slugworth]

I record on a Tascam dp-008ex and after mastering I occasionally use Audacity (open source freeware) for post-production. I am aware that those two items alone likely limit my hopes for better vocal production/separation/space or other term I am trying to relate.

I avoided recording through DAW and computer programs because I am concerned of the learning curve and the possibility of less instrument time due to more software diddling. However, I will admit that going over whole drum tracks instead of tracking bits at a time and editing them in as needed, gets tedious, frustrating and time consuming. I guess one upside is it forces me to play drums more than I normally would. I've been playing my wannabe E-drum kit Yamaha DD-65 for roughly eight months. I seem to be progressing more lately which I attribute to increased recording.

Anyway, back to my vocals concerns. I've watched YouTube videos on recording vocals. I have employed some tips with good results. Those tips mostly have to do with mic placement and insulation/isolation.
Seems like a lot of videos point to layering vocals for a fuller sound (most using DAW). Maybe I am old school, but I find it irritating when I can't distinguish one lead vocal from what seems to be four or five backup singers (in reality just layering) hitting the same notes. Yes, it sounds bigger, but also sounds too processed and a cover up for possibly a poor singer.

In saying that I would rate my voice average at best. As much as I would like my vocals/voice to sound better I don't intend to hide behind layering. Well, at least not layering overkill. So, how does one achieve wider sound when recording vocals. I've tried copying a vocal track on two channels and then panning each channel full-lock left and right. It helped a bit, but not to what I was hoping for.
The projection of my recorded vocals whether I listen back on monitors or headphone sounds as if I am speaking (singing) directly in front of a listener. Too "in your face" (for lack of a better description)

Any tips, suggestions, thoughts?

 

[-mjk-]

this topic in itself, could be the contents of the entire book. Copying a track and panning a left and right does nothing. But you might want to do however is take one track and vary it by 1 to 6 ms. in order to make it spatial. Remember though just because you double track doesn’t mean you have to use the double track at 100%. You can bring it up 10 or 15 or 20% so it just fills it in a little more in order to make it thick.

 

[mixerizer]

I would add - you can thicken / widen / double track all different kinds of ways, but the tastiest and easiest imho is when u really really have your vocal down, u know what u want to do w all the inflections and subtleties, and u nail the performance. Over and over.

Two extremely similar and compelling takes by the same vocalist, intended as a double, mixed to taste, will do the job. To see the difference, one exercise is to have the overdub only consist of the closing half of each stanza, to hear the accuracy without riding mute.

 

[mixerizer]

Also, since u r working in ur own studio and have time, u can practice singing along w the vocal as u listen to the rough mix, give it a little while till u know it cold, and then double track it. If it’s overnight or a few days, the difference in your voice between the two takes might also help create a blend of timbre.

 

[Slugworth]

this topic in itself, could be the contents of the entire book. Copying a track and panning a left and right does nothing. But you might want to do however is take one track and vary it by 1 to 6 ms. in order to make it spatial. Remember though just because you double track doesn’t mean you have to use the double track at 100%. You can bring it up 10 or 15 or 20% so it just fills it in a little more in order to make it thick.

Thanks for the reply. I haven't tried a 1-6 ms delay. I have tried, though not purposefully, a partial volume double track. Several months ago on one particular song I recorded double vocals onto a second track. All of the same lyrics sung in a lower pitch than the original track and with some mild echo and reverb added to it. Adding it in at full volume to the first track sounded worse than just one single track. A thought came to me to try it at a low volume (roughly 10%) which made it just audible to hear without being obtrusive. It did give the vocals more depth and width. Unfortunately, I didn't even like the primary vocals, erased all vocals and that song is an instrumental in it's current state. Another one I intend to circle back to for re-recording to tighten up timing. I will try that technique again though soon on another song.

 

[Slugworth]

I would add - you can thicken / widen / double track all different kinds of ways, but the tastiest and easiest imho is when u really really have your vocal down, u know what u want to do w all the inflections and subtleties, and u nail the performance. Over and over.

Two extremely similar and compelling takes by the same vocalist, intended as a double, mixed to taste, will do the job. To see the difference, one exercise is to have the overdub only consist of the closing half of each stanza, to hear the accuracy without riding mute.

I like the less is more approach and agree with the bit about inflection and subtleties, of which I am learning more about through vocal recording time. In the video below, maybe it's just me, but the original raw track sounds better than the finished version.

 

[Arjan P]

Thanks for the reply. I haven't tried a 1-6 ms delay.

In my experience, the copied and delayed vocal only works if you keep varying the amount of shift - and do it on the original and copy both. Lots of work, IMO not worth it against singing the same lines another time. But you need a singer who can do the exact same thinhg twice - not for everyone..

Several months ago on one particular song I recorded double vocals onto a second track. All of the same lyrics sung in a lower pitch than the original track and with some mild echo and reverb added to it.

But that's not a vocal double if you sing it an octave lower. Doubling means doing the exact same thing twice. and then you can either pan both versions away from each other (extreme L/R or just a bit L/R) or you can also work with EQ on the doubled take and keep it 'behind' the main vocal, both centered.

Some good examples of doubling:
Here, There and Everywhere - The Beatles (or me elsewhere in the forum ) - Extreme L/R
Something - The Beatles (same ), especially the parts 'I don't wanna leave her now etc..
Julia - The Beatles
Dear Prudence - The Beatles
And those are only some Beatle examples!

 

[mixerizer]

Arjen is 100% right, esp w the Beatles references. The ways they get a musically powerful result from their doubling and vocal hard-panning are downright astonishing.

 

[Slugworth]

But you need a singer who can do the exact same thinhg twice - not for everyone..
But that's not a vocal double if you sing it an octave lower. Doubling means doing the exact same thing twice. and then you can either pan both versions away from each other (extreme L/R or just a bit L/R) or you can also work with EQ on the doubled take and keep it 'behind' the main vocal, both centered.

 

[Slugworth]

I should clarify that I did hard pan the tracks L and R at full-lock (car steering term). I sang the R track with similar inflection as the L (original vocal) track, just a lower octave as mentioned prior. I've tried on a couple of songs in the recent past singing identically on a L and R track and it just doesn't sound good to me. Maybe today or Monday when I have time I will record just vocals and post a link here to hopefully find out what I doing wrong or could do better.

This seems to be a good vid on the subject

I like the mono double and stereo double techniques for lead vocals that he demonstrates. More layered than that sounds like crap to me. Or should I say sounds like The Weekend and every other pop singer that's over-produced these days.

 

[mixerizer]

You wanna talk vocal layering (vs doubling)?

when you are ready for pure, pure insanity check this technique out

 

[Arjan P]

I like the mono double and stereo double techniques for lead vocals that he demonstrates. More layered than that sounds like crap to me.

But layering is not the same as doubling, as @mixerizer points out. And the 'mono' and 'stereo' techniques are the ones The Beatles used in those examples.

 

[mixerizer]

And, for a Beatles example with comparatively loose / messier doubling, check out ‘Martha My Dear’

timing, timbre, emotion, all vary but the melody does not, it’s a unison overdub, trying to be close.

and it’s two vocal lead tracks (with live hand claps) panned right down the middle w the piano and strings left, guitar center-left, the bass center-right, horns and reverb right what I never noticed but learned from the notes below was that the drums and guitar are on the same track.

“on October 4th, 1968, Paul, Ringo and George Harrison entered Trident Studios in London at 4 pm to begin working on the song. The rhythm track was recorded first, the instrumentation being Paul on piano (track seven) and lead vocals (track five), with both George on electric guitar and Ringo on drums together on track six of the eight-track tape. Assistant Mal Evans, in "The Beatles Book Monthly," wrote that "Ringo bashed a hole in his brand new bass drum skin the night we started this track," which was undoubtedly replaced shortly thereafter. They may have run though it a few times without the tapes rolling but, in the end, this rhythm track was recorded in one take. This was complete by 6 pm in time for studio musicians to enter to record the score for his previously recorded rhythm track for the song “Honey Pie.”

Then, after this group of studio musicians left, another set entered, 14 in number, to record the orchestral score that George Martin had recently prepared for “Martha My Dear.” As the above photographic evidence suggests, these musicians recorded their parts in two shifts; six brass musicians with Paul and George guiding them on their instruments, and then eight string musicians recorded their part afterward. As for the first shift, the trumpets and horns were recorded on track three, while the trombones and tuba were on track four. The second shift then had the performance of all the string players put on track eight of the tape. By midnight, these overdubs were complete, musician Leon Calvert contributing both a trumpet and flugelhorn part in this arrangement.

As we see, this was a very productive day in the studio. However, it still continued. Paul kept producer George Martin and engineer Barry Sheffield in the control room so that he could record more overdubs, this taking the session to 4:30 am the following morning. Paul re-recorded his lead vocals on track five of “Martha My Dear,” erasing the vocals he performed during the rhythm track, adding handclaps during the instrumental section of the song. He then double-tracked both the vocals and handclaps onto track two of the tape.

Oct 5th, 1968: Paul took to overdubbing more onto “Martha My Dear,” this being his masterful bass guitar part, which was undoubtedly recorded onto track one of the tape. Mark Lewisohn, in his book "The Beatles Recording Sessions" suggests that Paul also recorded an electric guitar part for the song at this time. Since we know George was also present on this day, it could have been either Beatle. Nonetheless, since both Ringo's drums and George's guitar were mixed together onto the same track when the rhythm track was recorded on the previous day, we know for sure that George is indeed heard on the finished recording, although a guitar part from Paul may also be in the mix somewhere. This completed the recording of "Martha My Dear," resulting in George Martin and Barry Sheffield creating both a mono and stereo mix of the song, only one attempt needed to create each of these.“

http://www.beatlesebooks.com/martha-my-dear

pretty tight vocal doubling for the last take at the end of a day like that

 

[Slugworth]

Turns out one my fav songs has vocal doubling, I chose this particular vid of the song because it also has great bass lines.

The vid below helps clarify doubling for me

 

[Slugworth]

Mixizer. After listening to Marth My Dear. I now know how ELO's sound was born.

 

[mixerizer]

Hehe yep Jeff Lynne doing prob 4 overdubs, not just one talk about lush! It’s like a human chorus pedal. Enjoy!

 

[mixerizer]

And oh nice! Ian Anderson has such a singular style. I’m listening and it sounds like he does slightly different dips that and during the second verse syncopation is where u can hear differences in his doubling best great example.

 

[-mjk-]

The original question was how to widen vocals. The method that I proposed (which should be relatively easy to implement on the equipment available to the OP according to his profile) is to vary the playback of 1 track by a few milliseconds. That takes advantage of precedence effect and will indeed make virtually any source seem wider. Introduction of a variable amount of playback delay will make the vocal width change and may be useful to some.

Now the discussion has evolved into other aspects of vocal production and seems to now be focused on doubling. So here is my contribution to the discussion:

If you don't know what you are doing, doubling a vocal track can really make it sound bad. It requires quite a lot of practice and skill (quite frankly) and you can't necessarily expect it to be the magic bullet for vocals. I myself have decades of experience doing it and I will share my insights here.

I normally do not like the intimate sections of a song doubled. Typically at least the first verse, and maybe even every verse, I like to keep as a single vocal. When I remixed a very famous song, I omitted the doubled track on the verses to bring it back down and more intimate. You may judge the effectiveness of this for yourself. My basic rule of thumb is this: I don't want the doubling to be so obvious that it detracts from the song. When I remixed this song that @Mark Richards had also done, I eliminated the doubled track altogether because it was not tight and IMO drew too much attention to sounding like 2 people singing. You can compare versions in this thread.

I always triple track background vocals. (Not tooting my own horn) but you may want to listen to this tune I recorded in 1991. There are 5 vocal parts consisting of harmonies and passing tones and there are 3 tracks for each voice: Left/Center/Right. I learned this technique from engineers who were familiar with how the biggest British band in the world (at that time) did background vocals. The verses are single-tracked, but when the background vocals come in during the end of the second verse, it is literally a wall of sound. Sometimes I do double-track the lead vocal and sometimes I do not. It depends on how strong the performance is. It's not often that I get a chance to talk with a recording artist about their record but once I had the opportunity to talk with the lead singer of the band Texas and I told her that IMO her lead vocal was so strong that the doubling was detractive. She was in a a radio station for an interview and did an impromptu live performance in the studio and this woman is a monster. Very strong and precise singing that didn't need the support of doubling.

If asked I will tell how I did those triple tracked background vocals. Some of the hardware devices used are not easy to find but there are some plugs that can work in place of them.

But that's not to say that a doubled vocal indicates that the lead vocal is weak! It's all about the effect that you want to convey. It's also about not being over-the-top and drawing attention to the fact that the vocal is doubled.

There is a very very good example of how to implement doubling right here on @Arjan P's SoundCloud channel. It's there, it's subtle and it very much adds to the track. Notice that the vocal track is thicker. To my ear it sounds like the second track is about 3dB under the original track. The balance between the fundamental and the double is rather critical.

 

[Slugworth]

mjk. Thanks for the informative reply. I will likely apply your advice for trying out milliseconds delay.

I've never heard of the band Texas or their female vocalist. A big female voice that comes to mind is Heart's Ann Wilson. Recollecting, I don't think she used double tracking much, if at all. Probably little need to, especially when her sister Nancy could join in for great effect.

I should have time to record vocal tracks using the techniques discussed here. I intend it all to be done on the multi-track recorder. If I tried it with Audacity (if possible) it would likely take me long time and cause undue frustration.

 

[-mjk-]

Use whatever is the best for your situation. Texas is still around, but I met them when they were still young (late 80s). They are from Scotland if memory serves, so what's in a name? lol

For the precedence effect widening technique: duplicate a track. I mean to actually copy it so it is identical. Pan the original and the copy hard L/R (doesn't matter which/where). Start by nudging one track back 1 ms at a time until you find a sound you like.

Valhalla has 2 free SuperMassive plugins that are very good on vocals. One of them has a random doubler that when used properly, sounds pretty natural. I used it on Zane's lead vocal on Talk To Me Baby and you can't hear it working. I just thickens it up a bit.