I need a bit of mixing advice

29 posts / 0 new
Last post

To keep it short, I'm looking for ways to bring vocals forward in the mix without turning up the volume. I usually do a boost at 2-4000 Hz, add a bit of natural reverb and use a vocal compressor (if needed). But instead of a neat plate of meat, potatoes, vegetables and gravy, I end up with stew.

So, should I also do something about the piano, the guitars, the bass....? Or get a better mic than my Shure Beta 58a, or replace my crappy Behringer XENYX 1002FX mixer? Any advice would be much appreciated.

Compressing the vocal will help bring it forward. If it has reverb on it, use less reverb (reverb makes things sound farther away.)

But really the main tool is turning other things down to open space for the vocal. You could try cutting in that 2-4 kHz range on other instruments, or panning things way out to the sides to open up the center for the vocal, or just dropping the volume of every other track by just a hair until you start hearing a difference.

Deep topic, for me this is a lifelong learning project, and I'm not really very good at it yet.

I know absolutely nothing about mixing - I only know what I've stumbled across that works, so take all of this with a HUGE grain of salt please.

I saw this trick in a youtube video (I wish I could post a link, but if you search you'll probably find it) Use the pan knobs to move the instruments around. Apparently, the brain perceives the sound directly in front as the most important, so you move the other sounds to the sides, as if they were on a stage. I've had success with this. Doubling guitar tracks then spreading them out to the sides with the pan knobs and moving them to the back with compression opens that center up for the voice. I usually leave bass and drums where they are. It's mainly tracks that sound the same notes as the vocal line.

EQ on voice: I use a high pass filter on the voice too to cut out the muddy room noise and non-vocal frequencies, and sometimes a low pass filter too if I wanna cut some of the snakey sounds.

Also, compression on the other tracks. The fx ReaComp has a preset called "Move to the back" that compresses the frequencies in the vocal range, also clearing way for the vocal frequencies. Again, I'm talking out my ear on this, cuz I'm self taught, and this may be the complete wrong way to do it, but it works for me right now Smile Good luck!

yeah I was gonna mention panning harder to the left and right for the potatoes and vegetables while the meat stays centered with less reverb!

I suspect that you have a clash with guitars, piano, (drums?) and vocals in the same sonic space. In general, it is EQ, volume, and panning to try and fix these issues. There are phase cancellation techniques that can clean but too deep for me to get into here. Oh, and you probably could use some automation and or sidechaining compressors for ducking instruments while the vocal is active. I think that vocals always override instruments and have more priority to be heard. Therefore, I would try a combination of volume, EQ, and automation. If you have access to mid-side EQ then maybe make sure you don't have too much bass frequencies on the side channels (cut low frequencies on the side channels). This can help clean the stew and possibly be what you need. The last thing I've got today is to cut instruments. They don't really matter. What matters is the vocals, the lyrics, and the clarity of your song.

So so happy to see others saying the same thing I did. Means I really did learn something! Thanks for answering guys! :)And BithProd thanks for askin!!!

Like @brrse said, another useful tool is a low-cut EQ (called "High pass", because it blocks the lows and lets the highs through). The only tracks that need their low end intact art things like bass drum, bass guitar, maybe a piano or keyboard part if it's really central and there's not much else going on. Literally every other track could get everything below, say 100 hz cut (try different frequencies to see what works). It helps clean up the mix.

Generally speaking, most men sing in the 200 - 600Hz rang. (Yes it functions to a greater range, but suspect this is your issue range, C4 to C5 ish) which matches your functional instrument mid-range, sans bass - treble.

Many times a, e.g, guitar can be used to enhance a vocal timbre, but sometimes, gets muddy. Nothing in music functions well as a pure tone, versus harmonics working with it.

For your specific instruments, you'd need to EQ Notch ID the frequencies that overlap. So, example, you like bass in your vocal, IDg your Mic, with its, proximity effect adds bass ... You may have to live with a more airy vocal, and cut that aspect, allowing the bass, or guitar to live there, typically on MY instruments 50 - 100Hz where it gets wonky. You'd move your notches in .25 - .5 - 1db max increments, -- and use NEUTRAL, studio monitor Headphones (no bass coloring etc. "Monitors") to hear it with.

You'd then check the result on a miriad of speaker platforms, ear buds, speakers, car stereos. Count on loosing all bottom on phone, PC speakers, getting muddy in car stereos, or heavy bottom since most color it that way. Get generic earbuds and audiophile buds. Make notes, come back and hear what you have to do via you EQ Notches again, repeat... Typically, generally speaking, 1/2 doz times and you'll be in you OWN ball park range.

Remember, you are adjusting vol., but in as small a notch space as you have access to, and time and patience. Typically you start Flat in EQ, then, drop, do not increase it, here. Remember too, this is a dry track only, 1 sound source per track exercise.

I'd suggest that 1 - 3 hrs max per day, due to EAR fatigue. In a month of daily work like this and you'll be were you need to be. Also, get a db meter, and with a Fresh Ear, check the db level at the distance from which you listen and hear the presence you like to hear, headphones at cone, auto at highway speed window up, home stereo 6ft? 1/2 vol with tone set to neutral... As example. The db listening levels affect the presence, hi/lo Hz, as heard.

I assume you have normal hearing. If not adjust for that.

Then, in copied track, not original, add Reverb, etc. using leveler functions can bring up timbre overall. Generally, generally, track what you do first. For, e.g,, I might EQ ( on a good recording to begin with ) , then, Reverb controlling the MANY aspects I have control over room size, wet/dry out +-, Lo/Hi pass etc. there as well... , then compress the overall track controlling well, attack,release, compressions ratio, etc.

The L/R pan I adjust visualizing a stage setup, how might that be?, vocal center, bass R, guitar L, drum kit MR, etc. To much seperation makes it an effect, not spatially relevant, -- choose which you want.

Generally speaking, most are listening with cheap buds, in a phone, iPad, PC. Many will use headphones. Nothing matches the fidelity of a headphone... I shoot for those and decent, non-audiophile buds, for today's folks. I don't produce finished tracks ... But if did, would check auto, home stereo then at 1/2 vol with tone neutral.

I didn't get into if using WAV or MP3 ... WAV is far better and orders of magnitude bigger... Stick with MP3 unless have a good reason not to.

If you get extreme feedback, ASK specifically, on what, how listened, dup that if need to, don't take it to heart... And don't bother trying to explain... It's just not worth the pain of effort : )

Then, yoyo ... : )

Try this, it's general... nothing's that easy : )

I have no idea how long this may be there, so save it, if feel is useful. There are many around, different explanations, so many ad hoc variables.

Generally, the automated tweeky filters engage finite ranges, -- problem is, they don't tell you their algorithm ... but, then, tweeking after is far less work.

http://2.bp.blogspot.com/_hwnCNuE0Gqk/TUs7ODXE9lI/AAAAAAAAAKA/hvdGzxRHlIc/s1600/Interactive-Frequenc...

Cool @nutation! Thanks for the great post. Audio production is one of my favorite subjects. Smile

I did do a little bit of mixing last week. What I did was cut instruments and reverb (dry vocals), and wide panning on the instruments. They sound better to me. I think I said earlier this week that I am moving away from the commercial sound (radio songs). They just don't sound good to me anymore. That said, a man/woman with just a guitar and vocals is not very appealing to me either. In fact, I'm kind of tired of the human voice...lol. I don't know...people change. I really don't care what people have to say. It's strange, especially for somebody like me who has love popular music my entire life. Though I find myself enjoying classical music more than the other genres these days. What's up with that? They were great! Today's commercial music is here today gone tomorrow. I haven't heard anything great for a long time. It might be good but not great. The old school composers were very precise with their Sonatas. This crying in your beer and shouting down the system is pure crap, imho. Interesting too that Rock and Roll is based on sticking it to the man with attitude, and I use to love the stuff. Well, I still like popular music but verse chorus bridge (let's add intro and pre chorus too) is not exactly unique or timeless. Lyrics, however, can be timeless depending on what you say. If you say something that still makes sense in 200 years....I think you did good.

I'll try some of that notching you mention in my next mix. I'll also be cutting more. For example, not play guitar through the entire song. Maybe only bring the guitar in for the chorus. just stuf like that. Thanks.

Most of my advice has already been mentioned but I like to talk so I'm gonna say it anyway.

Instead of using reverb, try using delay on the vocal. You'll have to play with the settings, but if you can get a fairly quiet tale without too much feedback (too many repeats) it can sound better than reverb and no-one will be the wiser. Generally speaking, the more reverb the more distant the brain thinks the sound source is. Delay *may* have a similar effect but not nearly as much.

Which means, if you add a bit of reverb to the rest of the tracks and barely any or none (or you use a tasteful delay setting) to the vox track it won't sound as distant. You'll have to play with this as too much reverb will make your production sound dated (at least, imo.) The rule of thumb I use with reverb (at least to start) is to turn it up until I hear it and then back it off just a smidge.

Use EQ to your advantage. Again, Nutation and Brrrse and Standup all mentioned using EQ and I'd have to agree. I have found the best advice with EQ is to use cuts instead of boosts whenever possible (which is more often than not.) The first issue is masking. If you find the sweet spot of your vocal and do a 2-3 db cut in that range in ALL THE OTHER INSTRUMENTS (or at least those in the same frequency range) then do a slight 1-2 db boost on the vocal you'll find a subtle but very effective way to make the vocal sound clearer. The idea is to keep other tracks from masking the vocal. I tend to only boost the vocal track around the 5khz range for clarity. It changes with the mic and voice, but I start there.

I agree with the high-pass filter on vocals. I actually use it on most tracks even bass (sometimes) and bass drums (less often.) My standard practice is use the stock Reaper plug in ReaTune. I select the lowest range EQ then set it to high pass (which is the same as a low cut.) You'll see the EQ sweep change to a nice round cut at the bottom end. Then with all the tracks playing, I sweep that up until I hear a noticeable difference in the sound. Then I back it down (lower) until I no longer hear that obvious cut. Somewhere between those two spots is the sweet spot. The key here is to make changes while listening to the entire song instead of just one track.

And finally, Panning. I generally like to start with LCR (using only hard left, center and hard right) panning. But in practice I don't always follow that rule. Typically the more simple the demo the harder it is to use. I always try to balance the panning. So if there is an instrument on the left (say a guitar part) it will sound odd if there isn't one on the right (although it can sound cool, too.) Your bass line and vocal will almost always stay down the middle. Almost everything else can be panned.

I also double some instruments to get clarity in the vox. (Sounds like you'd do the opposite, right?) So in a thicker mix, say a rock track with rhythm guitars, lead guitar, bass drums and vox, I would play two rhythm guitar tracks and pan one hard left and the other hard right. To enhance the effect you'll want to purposefully use different sounding tones for each track. The bridge pup vs the neck pup. A strat vs a LP. Or even two different distortion types or settings. I'll do a narrow band sweep of EQ on both tracks and find their sweet spots. I'll boost the sweet spot slightly (no more than 3db) and then notch the opposite track in the same area. This will really make the guitars sound wide and in most cases more powerful. And as a result of them being wide, your vocal (or any other instrument in the middle) has less to mask it which brings it out just a tad.

Now let's say you have an acoustic guitar and vox track. Typically you don't have a problem getting the vocal to stand out, you just turn it up right? Well, I often employ the Haas effect on the accompanying instrument to make the vocal really stand out without being louder. Especially if it's a fairly weak vocal performance (in volume and dynamics.) Quick rundown: Double the guitar. Pan one track hard left and the other hard right. Pick one track and move it 10-30 ms. What you've done is added a delay to the entire track. One side is a very fast echo of the other. You can't hear it other than whichever side is behind will sound very quiet compared to the other. So you'll have to boost that side to match volume. Once the volume is matched you'll want to adjust both tracks together. So I often will render a stereo pair of the guitar track and use it for fader adjustments. Or if I'm close I'll just lock the two tracks together so when I move one the other moves as well. It's a great effect but it can sound obvious depending on how it is used. You can use this trick in other mixes as well. That rock track I mentioned above. Let's say you only have one rhythm guitar track but it's a pretty hot track. You can use the Haas effect to make it sound wider and more full and maybe not quite so hot. Or maybe you have a keyboard or synth track in a rock song. You want texture and not prominence. This is a great trick for that, too. And the bonus is that this has the obvious effect of hard panning and clearing out space for the middle instruments (most often, vox.)

If you have any questions or I haven't made myself clear, just let me know and I shall explain further. Like most skills, these are all very easy moves to make... once you know how.

Definitely lower the instruments down, pan them, and see that helps. Also get rid of any bass in the mix. This will also help to make the vocals sound bigger. I don't think it really is a big deal if the vocals aren't so upfront as long as the vocals are intelligible throughout the song. I don't think reverb will make the vocals stand out, a compressor will do the job better. So use a compressor or have the singer sing closer to the mic.

my biggest advice in choosing from the various advices here? listen to the different people's tracks who are giving advice, and see whose tracks exhibit the effect you are looking for, before you decide which advice to try.

Good advice, tsu. But remember, we don't always use all the skills in super fast demos during 50/90. So I would find a song and then ask specifically about that song to get a good result.

Another thought: try recording the tracks less hot if you are like me and tend to keep them towards the top of the meter. Digitally, well under -6db is fine. It will help the mix a lot. I struggle with this myself and it is an easy thing to do.

Sing louder! shout! what? I can't hear you!

Seriously, tho, all the advice from above is good. My stuff is usually all about the lyric, in a way, so I mix my vocal very up-front most of the time, and yeah, usually eq the vocal track so that below 100 hz is cut, add a fair bit of compression (which helps when i sing softly to make it more 'even') and sometimes let those high frequencies get a bit of a boost (or cut the other stuff) to make the whole thing more understandable. And compression on the track as a whole sometimes can bring up the volume of the music for the parts where the vocal isn't, if that makes any sense!...

As TCE says above, too...

-- Oh, careful using "here" as if advice is "good", (it is free though : ) (also, that's a great way to shoot the horse you're presently riding, too), e.g.,
-- David Gilmour, Pink Floyd, uses his phone. Actually he, in one anecdote, could not get the feel of what he sampled to self, so used it in released work (see some latest interview stream). Sounds awful as-is, and even when run with other. That got tweeked, carefully.

If you listen to bare/raw/dry tracks, demos, most readily available, (others), Jimi Hendrix, vocal, instruments, Rolling Stones in France, come to mind -- any recording *here "sounds" better! -- Just say'in, and was waiting for that one : )

Additionally, Neil Young, in a recent interview, as I feel IS how many do, can hear an AM Radio'd Demo, first 10 - 30secs, and know if like it, is good, etc. (me too). Is anyone posting Mastered work "here", to sample? Production advice, is different than capturing a decent demo with what's available, your voice, or krappey guitar, uke, etc. Onlineguage, is a tricky thing, on a Subway ride, listening between stops.

Also, again, their instruments, vocals, et al. Function in what Hz range? How, Monitored? So many variables. (Had a "friend" sample a mix, on his phone, -- forgot, not everyone uses his ear buds, etc. and wanted vocals up front... ahhh, his non-audiophile, buds, added so much Bass, any adjusted mix would sound like AM Radio when remixed on that listen.

Consider on what 3000 other folks here listen with? PC laptop speakers?

I have problems with making the vocal track sit in the mix too. I have a Rode M3 ( small diaphragm condenser mic ) but when I borrowed AKG Perception 420 ( large diaphragm condenser ) from my work place I noticed a big difference for the better.

Large diaphragm condenser microphones are usually recommended for vocals because they have better transient response. The sound is more detailed and there is more "presence". But these microphones are also very sensitive to room ambient sounds and you need a pop filter and often a phantom power.

Also it can be a hassle to find something that fits your voice and the sound you have in mind. And they are more expensive than Shure Beta starting from 300 euros or so.

I'll add a bit of advice myself: Use good monitors and/or headphones. I just bought a Beyerdynamic DT 770 Pro, and a new world opened up. I could suddenly hear all the nuances, and mixing got one heck of a lot easier. Only problem is, they're TOO good, so I still need to listen to songs on an mp3 player and a computer to make sure nothing disappears in the mix for people with cheap headsets from eBay. Wink

As @Klaus says, getting a large diaphragm condenser mic can make a huge difference to your sound. I'd highly recommend borrowing one (or buying one, if you can afford it). They don't have to be stupidly expensive; the Røde NT1-A comes in a pack with a pop shield, shock mount and an XLR cable for £150 in the UK, €189 in the EU and $230 in the US. It's a great deal and it's a very, very good mic for the price. I love mine - it's great for recording acoustic guitar, too. But yes, they are good at picking up external noise: birds singing, dogs barking and kids yelling have all ruined takes for me in the past week Smile

In terms of mixing, remember that your amp only has so much energy to put into the playback. You want to share out the audio spectrum between your tracks so that each gets the most effective share of the spectrum, and nothing's fighting too much with anything else. Your bass and kick drum take the low end, cymbals take the highs above 8kHz or so. I tend to keep vocals between 500 Hz and 2kHz, so anything else gets edged down in those frequencies. EQing electric guitar tracks to take out the mid frequencies works well in recordings, even though it'll sound weird when you hear the track in isolation. For live performance it's almost always essential to cut the mids on electric guitars, particularly for rock or metal. You can always tell folks playing gigs who are only used to playing in their bedrooms, where that doesn't matter - get them in a band without paying attention to their EQ and the sound just becomes mud. Your piano track needs EQing in the same way, because it occupies the same frequencies. Cut the mids.

One other technique for making a track stand out more which I haven't seen mentioned above is sidechaining, where you feed your vocal signal into a compressor on the track that's competing with the vocals for bandwidth. When vocals are present, they "switch on" the compressor, so that the competing track is automatically turned down - as if you were riding the faders on it to control volume.

These days, sidechaining has another use: giving the compressor a slow release and sidechaining the kick drum into a synth pad will give you the archetypal Wub-Wub-Wub compression pumping that EDM producers love...

I actually own a Røde NT1-A, but it adds noise to my recordings. It probably has to do with the phantom power of my crappy Behringer mixer, as I bought an insanely expensive shielded cable. Or what do you think it could be, headfirstonly?

The NT1-A has an insanely low noise floor (Røde use it as one of the thing's unique selling points), so I'd be *very* surprised if that's what it is - but there's always the possibility that you got a faulty one. If you know anyone with a similar large-diaphragm condenser mic, borrow it and see if there's any difference in your recordings. If there is, take the NT1-A back and get another one.

However, my suspicion would immediately fall on the Behringer mixer; they have a reputation for being noisy. They're cheap for a reason. I presume you've tried both of the XLR sockets on it to see if there's a difference?

Rather than splashing out on a better-spec mixer, you could try a non-Behringer mic preamp with phantom power (the ART Tube MP is about 40 Euros and gets good reviews). You'd need to switch off the mixer's phantom power, plug the mic into the preamp, and plug the preamp into the mixer (or, if you can, directly into the audio interface you're using with your DAW. This approach lets you adjust (i.e. boost) the gain on the signal from the mic before it gets to the mixer; I suspect the mixer will need the input signals running as hot as you can get them without clipping to minimise the audibility of the noise.

I use a TC Helicon "Mic Mechanic" preamp (currently going for roughly 120 Euros) in my setup. It provides phantom power and it also has a de-esser, compressor, EQ and a whole bunch of echo and reverb settings to play with. I'm really happy with the results I get from it.

I just bought a new studio mic, a Sterling ST51 - will try it out later today when I get the cord. This is an upgrade from a Digital Reference DRV100? So I'm confident my vocals will be improving! Bought a pop shield and a boom stand too. And even if the mic isn't fantastic, it LOOKS so COOL! Smile

Bump.....

Such a great topic of discussion.

I haven't read everything but mixing can be much easier if we stop cluttering our vocal lane. Therefore, my advice today is to let the vocal breathe by "stop playing" during critical parts of the vocals. Or simply play softer where vocals are. Oh, and yes cut those fill instruments (keys) when the vocal is playing. Try to think riff then sing riff then siing instead of playing guitar and other instruments all the way through the entire song. I wish I could give you an example but I don't have one handy. I'll bring this up again, and give samples of what I am saying when the 2017 season starts. Oh, one example I can think of off the top of my head, "on the day I was born (riff) the nurses all gather around (riff) they gaze through the wide window (riff) at the joy they had found (riff).

Here's another tip was given from Graham (the recording revolution). I haven't actually tried this trick as described but I've always used automation on the vocals. One thing he said that I found true is the volume was correct but you can't hear it. So what they are attempting to do is trick your mind into noticing the track in question by only boosting the transient. I've been experimenting these kinds of techniques for years. I think this is worth taking note, and worth the time it takes to watch the video.

https://www.youtube.com/watch?v=zIOoiiI6wmA

I enjoyed reading through most of this thread again. Thanks for the bump.

I've been watching mixing videos again. I have a lot of tutorials from Groove3 and Recording Revolution and other YouTube stuff. So what is the reason for bringing this up, you may be asking. These guys, oh man, and there plugins... Damn man. Why do they have to use plugins on every single track? I use to be just like them. I'm changing a lot though and find the music is "better" before we destroy it with all of our expensive or cheap plugins. I think all we really need as far as plugins is EQ, Compression, and probably Reverb. But those dudes have all of that on nearly every track. There trying to make it sound "radio ready". Guess what? I don't like the way the music sounds on the radio these days. Especially Country music. It sounds too loud. Not the volume....the mixing and mastering. I could have the radio on low and it sounds loud. Idiots. Just my opinion, of course. I understand people are listening in their cars and other loud places and need the music to be louder. But at what cost? Do you know that if you buy a CD today it is squashed? Yes, the master recording is squashed. Idiots. End of rant

Hi guys

Well, this time Graham (the recording revolution) is thinking like a 5090 music maker. I have mixed feelings about his lecture here. For years I have been making up songs and posting to this challenge. I always thought I was doing it the right way. But why did I think I was doing that right? I'll tell you why. Because I'm an audio guy and I mostly want people to hear how good I am with the audio. What happened was I was spending more time working on so so songs than writing good songs.

https://www.youtube.com/watch?v=8ITMLbZRH0Y

Ok so this advice may good for creative writing challenges like 5090. However, if you read this document (the link below) about songwriting technical blunders and compare what they say with what Graham says, well, somebody is wrong.

http://vickyjohnson.altervista.org/10%20Technical%20Blunders%20Songwriters%20MakeEdited.doc

I'm interested in what you think about mass production versus writing good songs?

I should point out that I have written "good" songs during my mass productions throughout my career in 5090. However, I think I wasted far too much time and effort on tunes nobody cares about. What do you think? Is there and answer or is it really all about simply having talent, or not?

I think that's an interesting document you linked to. I need to make the time to listen to the video to make the comparison. Maybe later this evening...

Hey jcollins... great points. This has been a debate for some time in my little corner of the world.

To my understanding, most all songs start on keys, or an acoustic guitar. You'll see videos of heavy metal guys with their dreadnaught banging it out.

So for "me", this year I moved to the Phone Voice Record function. I've compared that effort to produced stuff of a week long for one track. For me, and others, they were not listening to the "hi fi" aspects. They were listening to the melody, reading the lyrics to word smith -krappe, and if they liked my voice at all, or the style of music.

Hi fi folks would listen to grass grow, so long as had good low end thump with crystal highs.

People *love my voice, others don't think I can sing... *hmmm?

Word smiths read the dictionary well, but not poetry, or lyrics intended toward mind-pictures.

Most like a pleasant melody, but not if into death metal, or rock a billy.

So my conclusion was, there was no good conclusion.

If you like great stadium rock hi fi acoustics recordings... do it if it makes you happy.

I settled on that, like many others, I like my voice, and get great satisfaction singing my songs in real time. I don't pop in CDs of me to listen to. But, do like the songwriter raw tracks I have to listen to, -- great satisfaction reviewing them from time to time. My own voice tends to put me to sleep... so, never quite finish.

I guess I should formalize a sound room, and have a setup ready to go, for like the Dylan quote in your links. I may get back to that.

Also, if ever "signed", by a label, your recordings, no matter how good, likely have to be redone by a union shop... a wierd thing that may still apply. That easily puts a few tracks into the $20k range quick and can't be avoided. But, that may have changed. Stay out of Nashville? Not really problem for me, even when family lived there :p

Good stuff Smile , so glad to see this posted to think through, well.