Recording vocals with decent loudness and without clipping?

[fpoir]

I use Scarlett Solo 3rd Gen with an AKG C 535 EB. Even with 12db+ headroom the vocals clip.

Suggestions appreciated.

Edited October 29, 2022 by fpoir

[Byron Dickens]

Turn the input down?

[fpoir]

33 minutes ago, Byron Dickens said:

Turn the input down?

Then the vocal goes too soft to mix. I mean, I could mix everything really at a very low level and then normalize it, and I'll use that method if I have no other option, but I can't believe that's my only way to avoid the clipping.

[Byron Dickens]

Then turn everything else down too

[fpoir]

On 10/29/2022 at 3:29 AM, Lord Tim said:

There's really no advantage to absolutely belting the signal in a DAW because the noise floor will be too low to care about in practice. On tape, we used to have to try to get the most out of everything going in to end up with a clean recording.

I'd recommend what was said - turn everything down, then boost it all at the end if it's too quiet for you... or simply turn your volume up. Don't normalize, that's a destructive process. Use volume or gain adjustments or automation envelopes.

Don't forget that most commercial music is mastered too, so what people give the mastering engineer is likely as quiet as you're hearing, and it gets boosted up by the engineer. If anything, giving a hot mix to a mastering engineer is a bit of a bad idea because they want some headroom to work with.

If the vocal you're trying to record is super dynamic, other that suggesting working on both vocal and mic technique (move in closer for the quiet parts, move away for the belts), running a hardware limiter after the mic and preamp but before your soundcard input is a solution. Something like a Empirical Labs Distressor is a common choice here.

Thanks. Makes sense.

[Tezza]

That's a handheld condenser,  it performs more like a dynamic stage microphone but with better, wider sound capture. Great choice if you want to record vocals in a space that is not soundproofed or acoustically treated because like dynamic stage mic's it's off axis sound rejection is pretty good. I do the same thing with a Shure SM86 or using a fethead with a dynamic mic. I get a much better recording of acoustic guitar and vocals than using a "studio" condenser. With the condenser, I just get appalling room reverb and 101 noises in the background.

It sounds like you might be getting spiking from being too close to the microphone. The closer you get to the microphone the more variance you will get in the signal from hot to soft, which can result in spiking which presents as clipping.

I don't know what your shock mount and pop filter solution is with that mic but sorting that out can also help. You can use a shock mount with handheld mic's if you need to. I use the type they sell on ebay for pencil mics, handheld mics will also slide into it, they are cheap and work as they should.

Next thing is with the pop filter setup, as it is a handheld mic, the tendency is to sit on top of it with one of those thick pop filters on top of it, but you can treat it like a recording mic and put a clip on circular filter before it.

I have different set ups for different dynamic mic's, see what sounds best. You probably want to be 2 or 3 inches away for vocal takes or if you have good mic technique and excellent monitoring capabilities, with headphones, you can sit on the mic more, depending if your a belter or more intimate singer. But generally, as you move back, the waveform will flatten more meaning less spikes (providing you can hear yourself in headphones).

The other thing is to have some sort of compression applied to the signal before it hits the DAW. This can make a big difference both in balancing the recorded signal and also in how you hear your vocal through headphones. How you can do this with the scarlett, I don't know. I use a Steinberg UR44 which allows me to do it.

 

Edited October 31, 2022 by Tezza

[Tezza]

Also, most of the information you will find on the internet about recording vocals is from the perspective of engineers, studio owners or non vocalists. They will say things like use a studio condenser microphone and compression is not that important in a 24 bit recording environment etc. This might be true but is very often not practical or desirable from a vocalists perspective.

If you don't have a decent "studio" to record in, you will get better results from dynamic/condenser handheld microphones rather than an expensive studio condenser.

For moving vocal recording to a more professional level, to create a great sounding environment for the vocalist, access to an excellent monitoring solution that has options for compression/reverb/eq is essential, not optional. Also, being able to balance volumes between your vocals and the recorded music is essential.

It's worth spending some time to set it up. If you can use shockmount, pop filter, mic distance, mic technique and audio interface input volume to prevent clipping on the audio interface, then there may be options in Cakewalk for setting up compression, reverb and eq for your headphones. I can do this in Cubase but I forget how to do it in Cakewalk so someone else might be able to help out there. I think you just set up an FX or AUX with the input from the main vocal track and then put the effects on the FX/AUX track and run it back out to your headphones either mixed in with the main vocal or without any main vocal.

 

 

 

 

 

Edited October 31, 2022 by Tezza

[Kal Elle]

On 10/29/2022 at 2:29 AM, Lord Tim said:

Don't normalize, that's a destructive process.

Is it destructive in Cakewalk only or in other DAWs too?

Reaper for example has this normalize waveforms feature that works great, but if it's really destructive I may have to quite using that.

And when you say destructive, you mean it messes up the quality of your audio?

Edited November 1, 2022 by Kal Elle

[Alan Tubbs]

This is where a hardware compressor/limiter helps.  I generally record thru a RND Portico II channel.  The compressor gets fast.  Not 1176 fast but enough to damp vocal pushes.  A gentle slope afterwards.  That works to firm up most signals but not so much that there is nothing to work on in the box.  An 1176 will still be able to light switch on/off the envelope or the slow changes wrought by an optical limiter.