Helix/Home Recording (Reaper) fun.....need help with vocals.

[pipelineaudio]

My favorite REAPER De-esser (if there's time) is the new spectral peaks view and I have it set so esses and cymbals are blue-black. Then a macro for cut at time selection/nudge -3dB

For the de-popper stuff, you may be right that its going nonlinear on an input and won't necessarily nicely be fixed in software, but for free with REAPER you could sure try a band of ReaXComp to control that issue

 

[rd2rk]

This is one definition, however, the "direct monitoring" on most interfaces is actually post converter latency.

I contacted Focusrite (I have an 18i20) and they confirmed that the conversion takes place when the signal is sent over USB, not at the Input, and is converted back when it returns over USB, not at the Output.

Not the same in what way? Yes, less in quantity, but both ARE latency, so the whole audiophile shakti stone green paint on the CD cable lift claims about ANY latency are just garbage out the window

If your beef is just that some people confuse 2-3ms with ZERO, then TECHNICALLY you are correct.
However, for most humans, 2-3ms might as well be ZERO, as it is (to most humans) undetectable.
If I call that VIRTUALLY ZERO, will that work for you?

I'm not stating it, the vast majority of recording situations going on in the world right now are stating it. EVERY SINGLE pro tools session EVER also proved it

Here's my statement that you responded to:

"You SEEM TO ME to be IMPLYING (if not outright stating) that millions of singers, recording in professional studios, find the latency experienced when tracking AND monitoring concurrently through the DAW to be totally acceptable."

You replied:

"EVERY SINGLE pro tools session EVER also proved it".

Proved what?
That every single vocalist who ever recorded in ProTools was totally happy to monitor their vocals while tracking with a 13ms delay?
Not buying it. I certainly wouldn't be happy with it, and I'm not NEARLY as picky about stuff as most of the vocalists I've known.

I'm sure that you're as tired of this discussion as I am, so here's (I hope) the last thing I'll need to say:

If you're a vocalist, and you can deal with hearing yourself with a 13ms delay in the cans while tracking, more power to ya!
Go ahead and monitor through the DAW and I hope your record sells a million copies!
Myself, I can live with 2-3ms latency to get a little reverb going, but much more than that is annoying.
That is all!

 

[SwirlyMaple]

Your setup in the Helix sounds very similar to what I am doing. I have an EQ > Tube Preamp > Retro > LA2 Comp.

I suggest moving the Retro Reel after the compressor. The reason is that if you hit the Retro Reel with too much signal, you get (realistic) tape distortion artifacts that aren't particularly pleasing to the ear. If you run the compressor first, it will make the levels into the Retro Reel more consistent, which avoids the harsh tape distortion and might also let you dial up the saturation parameter a little bit more. Also, I suggest turning down the Texture parameter in Retro Reel to a pretty low value for vocals.

As for thickening vocals, I've done a few passes that are double takes, do I just need to sing better to get past any sort of chorusing between the two? Or is that a product of how I have them panned/mixed? The one song I'm using multiple takes on so far is Trippin On A Hole In A Paper Heart. I have 3 vocal parts going, the first line of the verse, 2nd line of the verse and the chorus. Each of those vocal parts I've done a 2nd take and have messed with some panning a bit. I Know its hard to fix without hearing, but I wonder if a 3rd take would be better in the center and then filling with panned L/R takes as needed??

Recording vocals is where a person finds out what a skill singing really is Yep, for layered vocals to not sound overly "chorusy" and just weird in general, they need to be sung consistently and accurately. (Unless you're intentionally going for a chorusy vocal sound, in which case the more differences there are in the takes, the better.)

Here's my usual approach to vocals:
(1) Have 1 vocal track be your main, centered vocal. Don't pan it. Try to make this one as accurate and perfect as possible, as it's going to be carrying the vocals in the mix. This one should have the most compression on it to keep it forward in the mix.
(2) Record 2 more takes of the same, and pan them about 20% left and right of the main centered vocal. Pull the levels on these two down. You want them to be "filling" the main vocal, but it shouldn't sound like 3 separate voices. I like to set the levels on these to where it's not obvious they're doing much of anything, until you mute them, when you'll notice how much thinner the main vocal gets.
(3) For width, you can do two more vocal tracks and hard-pan them about 70-80% or more left and right. Again the levels on these should be quite low. You want them to add to the sense of spaciousness but not be standing out above anything else. Extra reverb on these two will really broaden the sound, e.g. for a big wide-sounding chorus repeat.

A few other tips:

• EQ is really important for mixing vocals. A small bump or cut of 3db or less in the right spot can be the difference between muddy vocals and something that sits well in the mix. It's natural to want to just increase the vocal levels until you hear them above everything else, but that's not a good approach. It's better for them to have an eq range carved to fit them in the rest of the mix, and then (if necessary) slightly boost the prominent frequency of the vocals. This will make them clearly audible while also allowing their volume to be pulled back, making the entire mix louder while still having vocal clarity. This video has some good advice about learning to hear this: (That guy's youtube channel also has a LOT of good mixing advice -- it's worth watching.)
• With dynamic mics, you don't want to get too close to them when recording vocals. About 8-12" away is a good starting sweet spot. The flub and boominess you get from proximity effect may sound fuller to your ears when you record, but it'll just sound like flub and mud in the mix and won't do anything to help them stand out. What we perceive as vocal power in well-mixed tracks is actually a *lot* thinner than you'd think. With a singer like Scott Weiland, his vocal fullness and cut comes from his excellent resonance and a unique distortion harmonic in his voice that floats above the sung note. Singing technique is a whole 'nother topic in itself, but in short: vocal "power" in a mix does not come from volume, either as it was recorded nor in the mix. It comes from vocal resonance (raise that soft palate!), and a clear/cutting tone. We sadly can't all be Scott Weiland (who IMO was one of the greatest rock vocalists ever and had a gifted unique voice), but we can all get a reasonably good, cutting resonant tone with the right technique. A vocal sung quietly but with a really crisp resonant tone will cut in a mix much better than a shouting vocal that lacks resonance.

P.S. Not sure what your level of singing ability is, but if you're looking for good advice on how to get closer to the way legends like Weiland, Stayley, and Cornell sounded, check out "Bohemian Vocal Studio" on youtube. He is the best I've found for really explaining and conveying the importance of vocal resonance, vowel shaping, etc. to sing with cutting power and without strain. Notice how relaxed he is in his upper registers and how little air he's using, but how much edge and cut his voice has. I think it's easy to assume those singers were just screaming their lungs out since they lived so recklessly, but they all had extremely good vocal technique at their best that allowed them to do what they did so well.

 

[pipelineaudio]

I contacted Focusrite (I have an 18i20) and they confirmed that the conversion takes place when the signal is sent over USB, not at the Input, and is converted back when it returns over USB, not at the Output.



If your beef is just that some people confuse 2-3ms with ZERO, then TECHNICALLY you are correct.
However, for most humans, 2-3ms might as well be ZERO, as it is (to most humans) undetectable.
If I call that VIRTUALLY ZERO, will that work for you?



Here's my statement that you responded to:

"You SEEM TO ME to be IMPLYING (if not outright stating) that millions of singers, recording in professional studios, find the latency experienced when tracking AND monitoring concurrently through the DAW to be totally acceptable."

You replied:

"EVERY SINGLE pro tools session EVER also proved it".

Proved what?
That every single vocalist who ever recorded in ProTools was totally happy to monitor their vocals while tracking with a 13ms delay?
Not buying it. I certainly wouldn't be happy with it, and I'm not NEARLY as picky about stuff as most of the vocalists I've known.

I'm sure that you're as tired of this discussion as I am, so here's (I hope) the last thing I'll need to say:

If you're a vocalist, and you can deal with hearing yourself with a 13ms delay in the cans while tracking, more power to ya!
Go ahead and monitor through the DAW and I hope your record sells a million copies!
Myself, I can live with 2-3ms latency to get a little reverb going, but much more than that is annoying.
That is all!

When you decide where and how far to shift the goal posts, I'll deal with your No True Scotsman

 

[SwirlyMaple]

When you decide where and how far to shift the goal posts, I'll deal with your No True Scotsman

No true Scotsman, or appeal to purity, is an informal fallacy in which one attempts to protect their universal generalization from a falsifying counterexample by excluding the counterexample improperly.[1][2][3] Rather than abandoning the falsified universal generalization or providing evidence that would disqualify the falsifying counterexample, a slightly modified generalization is constructed ad-hoc to definitionally exclude the undesirable specific case and counterexamples like it by appeal to rhetoric.[4] This rhetoric takes the form of emotionally charged but nonsubstantive purity platitudes such as "true", "pure", "genuine", "authentic", "real", etc.[2][5]

Example:
”Almost every recent recorded vocal has 13ms monitoring latency and it isn’t a problem.”

“I and many others find it to be a problem. Studios have ways to avoid it.”

“That’s just your preference. It’s not a problem for actual singers.”

Edit: next we can cover your claim of “false dichotomy” where no dichotomy was even present, if you like.

 

[pipelineaudio]

None of that was a no true scotsman, Try again. At this point, people are being wilfully ignorant. There is a Alexandria sized library of peer reviewed published research on this subject. I'm just the messenger and not at all interested in wasting time against motivated fallacious reasoning. Just trying to save the OP from bad and ignorant advice

 

[SwirlyMaple]

I'm just the messenger and not at all interested in wasting time against motivated fallacious reasoning. Just trying to save the OP from bad and ignorant advice

You mean like all the lengthy posts of help I’ve provided in this thread?

I remember providing hours of help to you on a recent occasion, too. IIRC it involved you locking up your device repeatedly with an endless-loop MIDI loopback, which you blamed on Line 6 and made an entire thread admonishing them for it, claiming the firmware was unusable and full of bugs. The best part was wasting all of my time when you insisted you had disconnected everything (the very first thing I suggested), only to reveal the next day that you still had the MIDI loopback plugged in. Talk about fallacious reasoning…

 

[The_Kid]

I suggest moving the Retro Reel after the compressor. The reason is that if you hit the Retro Reel with too much signal, you get (realistic) tape distortion artifacts that aren't particularly pleasing to the ear. If you run the compressor first, it will make the levels into the Retro Reel more consistent, which avoids the harsh tape distortion and might also let you dial up the saturation parameter a little bit more. Also, I suggest turning down the Texture parameter in Retro Reel to a pretty low value for vocals.

Recording vocals is where a person finds out what a skill singing really is Yep, for layered vocals to not sound overly "chorusy" and just weird in general, they need to be sung consistently and accurately. (Unless you're intentionally going for a chorusy vocal sound, in which case the more differences there are in the takes, the better.)

Here's my usual approach to vocals:
(1) Have 1 vocal track be your main, centered vocal. Don't pan it. Try to make this one as accurate and perfect as possible, as it's going to be carrying the vocals in the mix. This one should have the most compression on it to keep it forward in the mix.
(2) Record 2 more takes of the same, and pan them about 20% left and right of the main centered vocal. Pull the levels on these two down. You want them to be "filling" the main vocal, but it shouldn't sound like 3 separate voices. I like to set the levels on these to where it's not obvious they're doing much of anything, until you mute them, when you'll notice how much thinner the main vocal gets.
(3) For width, you can do two more vocal tracks and hard-pan them about 70-80% or more left and right. Again the levels on these should be quite low. You want them to add to the sense of spaciousness but not be standing out above anything else. Extra reverb on these two will really broaden the sound, e.g. for a big wide-sounding chorus repeat.

A few other tips:

• EQ is really important for mixing vocals. A small bump or cut of 3db or less in the right spot can be the difference between muddy vocals and something that sits well in the mix. It's natural to want to just increase the vocal levels until you hear them above everything else, but that's not a good approach. It's better for them to have an eq range carved to fit them in the rest of the mix, and then (if necessary) slightly boost the prominent frequency of the vocals. This will make them clearly audible while also allowing their volume to be pulled back, making the entire mix louder while still having vocal clarity. This video has some good advice about learning to hear this: (That guy's youtube channel also has a LOT of good mixing advice -- it's worth watching.)
• With dynamic mics, you don't want to get too close to them when recording vocals. About 8-12" away is a good starting sweet spot. The flub and boominess you get from proximity effect may sound fuller to your ears when you record, but it'll just sound like flub and mud in the mix and won't do anything to help them stand out. What we perceive as vocal power in well-mixed tracks is actually a *lot* thinner than you'd think. With a singer like Scott Weiland, his vocal fullness and cut comes from his excellent resonance and a unique distortion harmonic in his voice that floats above the sung note. Singing technique is a whole 'nother topic in itself, but in short: vocal "power" in a mix does not come from volume, either as it was recorded nor in the mix. It comes from vocal resonance (raise that soft palate!), and a clear/cutting tone. We sadly can't all be Scott Weiland (who IMO was one of the greatest rock vocalists ever and had a gifted unique voice), but we can all get a reasonably good, cutting resonant tone with the right technique. A vocal sung quietly but with a really crisp resonant tone will cut in a mix much better than a shouting vocal that lacks resonance.

P.S. Not sure what your level of singing ability is, but if you're looking for good advice on how to get closer to the way legends like Weiland, Stayley, and Cornell sounded, check out "Bohemian Vocal Studio" on youtube. He is the best I've found for really explaining and conveying the importance of vocal resonance, vowel shaping, etc. to sing with cutting power and without strain. Notice how relaxed he is in his upper registers and how little air he's using, but how much edge and cut his voice has. I think it's easy to assume those singers were just screaming their lungs out since they lived so recklessly, but they all had extremely good vocal technique at their best that allowed them to do what they did so well.

Appreciate all that information. The singing technique, you're right, is a whole separate topic.

I sat this morning and fiddled with my Helix preset and made a new fx chain in Reaper. I'm happier with it. The one thing I need to do is bring up the EQ's of the guitar/bass/vocals so I can see where they are competing a bit and make some room for them. I did that on their own, but need to do that with the mix so I'm not 'layering' instead of 'blending' to get an overall mix, if that makes sense.

Right now, my vocals sound better than the first run, but while they are 'loud enough' vs the instrumentation, they do seem to be floating on top of the music and not integrated into it. I think EQ'ing a bit will help them lower the level and sit in better like you said.

My changes: I raised the input trim on the Helix XLR input so it was a touch more sensitive and I didnt have to 'eat the mic' to hear things (eliminated that proximity effect a bit), also added the Dynamic Room Reverb to the Helix preset as I didnt care for the Reaper ones as much. I'd rather 'post' it, but it seems easier to control in the Helix. I will check the Retro/Comp order and see how that works.

 

[SwirlyMaple]

Appreciate all that information. The singing technique, you're right, is a whole separate topic.

I sat this morning and fiddled with my Helix preset and made a new fx chain in Reaper. I'm happier with it. The one thing I need to do is bring up the EQ's of the guitar/bass/vocals so I can see where they are competing a bit and make some room for them. I did that on their own, but need to do that with the mix so I'm not 'layering' instead of 'blending' to get an overall mix, if that makes sense.

Right now, my vocals sound better than the first run, but while they are 'loud enough' vs the instrumentation, they do seem to be floating on top of the music and not integrated into it. I think EQ'ing a bit will help them lower the level and sit in better like you said.

My changes: I raised the input trim on the Helix XLR input so it was a touch more sensitive and I didnt have to 'eat the mic' to hear things (eliminated that proximity effect a bit), also added the Dynamic Room Reverb to the Helix preset as I didnt care for the Reaper ones as much. I'd rather 'post' it, but it seems easier to control in the Helix. I will check the Retro/Comp order and see how that works.

Good call on raising the input trim of the mic! I should’ve mentioned that—sorry about that. You definitely don’t want to feel it it’s taking extra effort to be heard when recording vocals.

One other thing that has helped me a lot is listening to the multitrack mixes of songs I’m trying to emulate, which lets you hear how the original vocals really sounded, how the tracks were EQ’d, how much reverb was on everything, etc. I’d probably be in violation of the rules here if I told you exactly where to get them, but if you search google for “multitrack mogg,” there is a blogspot domain that still has a bunch of them available. There are STP songs in there too (can’t remember if they include Trippin on a Hole, but there’s a good chance). These multitrack files came from the Rock Band series of video games, which had the original studio stems of *many* well-known songs stored in them. They are a gold mine for people wanting to hear how the artists sounded in the studio, and how the mixes were approached.

If you do get some of these, you’ll also need Audacity (a free audio program) to open them (unless Reaper can open them directly—not sure about that). From there, you can export each track to a separate file and load them into your DAW. Occasionally they contain a stereo track or two, but the rest will be mono tracks that come in pairs which need to be hard-panned left and right to restore the original stereo mixes. (i.e. find the two guitar tracks, and hard pan one left and the other right. Then find the two vocal tracks, and do the same, and so on.)

 

[MrTAteMyBalls]

Been there, done that, annoying.

I haven't spent much time recording, but I can't see why ANY singer would rather hear a delayed version of what they're singing vs monitoring it direct in real time.

I know there's LOTS of folks on here who have LOTS more recording experience than me.
Is this actually common? Vocalists preferring to hear their vocals delayed vs in real time?

With direct monitoring you cannot build up a channel with comp, delay, reverb....etc unless you have an interface with DSP or some outboard gear. Most people don't have that and a LOT of singers hate to monitor their dry vocal.

That's the reason. It's not that they PREFER to hear latency....it's that a little latency with reverb is preferable to a totally dry vocal with no latency.

 

[pipelineaudio]

as shown upthread though, you can run some parallel fx without too much pain, including a reverb

That's the reason. It's not that they PREFER to hear latency....it's that a little latency with reverb is preferable to a totally dry vocal with no latency.

But yes, finally someone who gets it rather than just going and looking for the worst part of what someone might be saying

 

[rd2rk]

With direct monitoring you cannot build up a channel with comp, delay, reverb....etc unless you have an interface with DSP or some outboard gear. Most people don't have that and a LOT of singers hate to monitor their dry vocal.

That's the reason. It's not that they PREFER to hear latency....it's that a little latency with reverb is preferable to a totally dry vocal with no latency.

I get that. But OP specified that he has a Helix. There's no NEED to do any of that.
The tools in Helix are excellent. It allows a DI track to be recorded concurrently.

Again, if a singer PREFERS the tools in the DAW and isn't bothered by the latency, all well and good.
I would definitely be bothered.

 

[MrTAteMyBalls]

I get that. But OP specified that he has a Helix. There's no NEED to do any of that.
The tools in Helix are excellent. It allows a DI track to be recorded concurrently.

Again, if a singer PREFERS the tools in the DAW and isn't bothered by the latency, all well and good.
I would definitely be bothered.

That's pretty cool actually. I have a Helix but I have never thought of using it that way. I guess with the USB connection you can always capture the dry track while monitoring the wet one??

 

[SwirlyMaple]

That's pretty cool actually. I have a Helix but I have never thought of using it that way. I guess with the USB connection you can always capture the dry track while monitoring the wet one??

Yes, this is one of the many ways it’s intended to be used. By default USB channel 7 is the dry guitar signal and USB 8 is the mic dry signal, but you can change the channel assignments in the settings if you wish.

You can use the hardware to direct-monitor your guitar or your vocals (or both at the same time), and have your DAW record the wet tracks and/or the dry tracks. Then you can either re-amp the dry tracks with the hardware or use Helix Native to do the same right within your DAW, as the plug-in has identical functionality as the hardware.

Those Line 6 folks, they’re pretty smart. And thank goodness they understand the value of low-latency, instead of waging the “Great Crusade of Latency Doesn’t Matter Except When It’s Your Preference Or I Say So But Regardless It’s Ignorant Advice To Suggest That Latency Matters.”