Somebody to Love Cover.
Diego
Moderator, 2.0 PRO, 3.0 Streaming Posts: 1,157
Okay, so here we have on of the tougher Queen songs.
I love this tune definitely one of my favorite Queen tracks.
@videoace @highmtn @Furious_Phil
Feedback is appreciated.
I love this tune definitely one of my favorite Queen tracks.
@videoace @highmtn @Furious_Phil
Feedback is appreciated.
Comments
I don't want to appear to be too critical, because I appreciate the effort you're putting into learning the parts and doing them. But there is a thing called "ear fatigue" that happens with mixes and recordings that have ragged sound in them. So you still have lots of trial and error, along with critical listening, to get your ear trained enough to clean up the recording and mixing aspects of your demos.
Once you do get those hurdles jumped, it will showcase your voice better, and allow people like myself to be able to hear more detail in what you are presenting.
Some people bury the vocals in the mix because they are hiding under the music. Others just lack the experience or the skills in putting the vocals where they belong.
One thing I learned long ago was that I had to identify the weakest elements in my recordings, and be able to present it in the strongest possible form that I want it to appear as in my mix. That might be the Kick, a Tom, the Lead guitar solo, Bass Guitar, the background vocals, the snare, or the lead vocal. You always want to be able to get the most important things dominant and crystal-clear before you start muddying things up with more and more elements. The lead vocal is the main thing people want to hear plainly in any recording. Everything you add affects everything else, and there is only SO much room in audio for everything to fit together. Exceed that, and you have mud.
Vocals are a tricky thing to record and have sound great on the playback. It's an art form, and very subjective. But certain things always have to be up to certain standards, regardless of the artistic interpretation. Some big records have intentionally distorted vocal tracks, just to make it sound that way. Microcuts comes to mind. I personally have a hard time listening to that, because, to my ears, although it's artistically the way the artist wanted it, it's very fatiguing to listen to. But I think that was the idea. Most people want clear, tight, clean vocals with a little bit of flavor added and maybe a little bit of edge or natural vocal distortion, up to a point.
What kind of microphone are you using?
How are you recording?
How are you mixing your vocal over the backing track? (i.e. are you just throwing it into Audacity and then trying to adjust the volume or are you actively trying to 'mix' it by using EQ and compression, etc?)
If you're more comfortable working yourself, as someone who's self-taught and knows a million was to screw this sort of thing up, I can offer the following advice re. recording your vocals:
- Get a decent microphone. You can pick up a USB one that'll do the job for about £50 (roughly $100USD). Don't waste your time recording through a potato...it'll give you a terrible level of feedback on how your voice actually sounds. I started out with a £1 desktop mic from Asda (WalMart) but I was only really doing weird experimental stuff, so I could get away with it. When I needed to do REAL vocals, it was only when I got a proper dynamic mic (a Shure SM58) that I could record a decent track. Even then, it still didn't give me the range of response I needed for my voice so I eventually upgraded to a fairly cheap condenser mic which I still use today.
- Maintain your distance from the mic/pop shield when doing different takes. Unless I'm very much mistaken, you've patched this together from multiple takes. There's nothing wrong with that at all, but if I can hear the inconsistency then it must be pretty obvious to more trained ears as I'm basically an amateur. Try to keep everything - volume, tone, proximity, etc - the same on each take as much as you possible can. It'll never be identical but it should at least be close because you want to create the impression of one continuous take. Like I said, there's nothing wrong with patching together takes - Barbara Streisand famously pieced a whole song together word by word to capture the performance properly, so this isn't a slur on your abilities as a singer; it's purely from an engineering and mixing point of view so to try to maintain consistency.
- With a voice like yours, you can safely cut all the frequencies below around 90kHz. This will free up a lot of headroom in the mix and also take out any rumble from the mic stand, etc. Be careful boosting higher frequencies when you've got as noisy as recording as yours appears to be as it'll just make it sound hissy and horrid. This is usually what causes what Bob's mentioned as "ear fatigue"; the ear literally gets tired out and can't hear higher frequencies as clearly.
- If you're doing multiple vocal parts yourself, learn to use panning and EQ. When you're mixing, you're creating a virtual space with sounds so think about how you'd hear multiple voices at once. You don't usually hear them all as if they're coming at you in a straight line, do you? No, they appear to be coming from a three-dimensional space and involve a mix of left, right and of both together. If you pan them TOO far, they'll poke out of the mix too much so maybe aim for between 50-75% on either side. With regards to EQ, try to EQ each vocal slightly differently and carve out a sonic space, otherwise they just must together which is what's happening on your track.
These are just a few suggestions, but like I said, feel free to drop me a line and I can try to do a better mix for you.
What is your recording set up? mic?, software?
Peace, Tony
I'm still a complete noob for recording stuff, so I tried experimenting around with the volumes, and it came out inconsistent. By the way Tommy, I JUST got a Blue Yeti USB microphone, I messed around with the settings for the mic. I want to know what's the most optimal set of settings for this mic for recording? Where can I send the vocal track?
I have my microphone in a separate table and I tried lowering the volume of the microphone but It didn't seem to work. But like I said, I just bought the microphone so It's a matter of experimenting with it.
Can this track still be fixed? I tried to get away from the microphone on some parts because I thought it was sounding either too loud, but apparently it still made it loud. On the verse where I do the C5, I had to do like 7 takes and every take it sounded to loud, and tried lowering the input to about -18 dB and it was still loud. I don't have a pop filter or anything if that's what's causing the problem.
Anyways thank you for the awesome feedback!
-Diego
You want the strongest signal possible going into the recorder. You can always turn it down later to suit the mix.
- 18db should still give a good signal, but you have to be closer to the mic. (I usually let my tracks "peak" at -05db)
As far as fixing what is there. No, and that's why you want to have the strongest signal going in. It doesn't matter whats going on with the other tracks. When you record a track that's the only signal you have to worry about. After you get all of your tracks recorded at their strongest signals, then you can play with it any way you want because everything will be strong, and clear.
Peace, Tony
Remember, a pop shield is only really there to reduce plosives (the p and b sorta sounds) which usually jump out of the mix in the lower end of the voice. There are ways to reduce them naturally but that's another story entirely. Try to stick to a thinner material like stockings rather than a thicker material as that'll interfere with your higher frequencies and deaden them. Thinner material, as you'd imagine, allows more information to go through it, basically.
When it comes to volume on the input, I'd suggest keeping it closer to a maximum of -2dB. Taking it as high as -0.5dB at the stage of the mix leaves very little headroom to place the vocal in a mix as it'll likely sit far higher than the original track. Nowadays, most people push for more volume so an overall mastered track sitting at -0.5dB isn't unusual, however I would argue, especially for audiophiles (or just annoying, perfectionist assholes like me), that aiming to have your final mix no louder than -2dB preserves much in the way of dynamics. A voice like Freddie Mercury's, were it compressed to the levels we see today, wouldn't have the same beauty and dynamics, particularly on a song like "We Are The Champions" or something. It'd sound loud and awesome, of course, but personally I feel it'd lose a lot of those beautiful idiosyncrasies he brought to the sound.
If you're recording using a USB mic, go into your main audio properties and look at how 'hot' the basic signal is when it's coming in. If you're seeing it hitting the red, it's too loud - Even the yellow area is too loud so try to stick around the peak of the green levels (assuming your computer shows you a coloured bar, but if not then try not to allow the levels to go beyond around two-thirds of the overall volume as it's coming in).
Try just using something like Audacity, a free audio editor, and import your backing track as a separate STEREO track (this is important). Set up a MONO track for your microphone (again, this is important for when it comes to the mix stage) and only use mono tracks for your vocals. They can be grouped together in stereo groups (like having the lead vocals separate, but your harmonies occupying their own space in the stereo field then grouped together to balance as an instrument, if you get me) once they're all panned and sat nicely in the mix, but that's something to consider later. By keeping the vocal in mono at the recording stage, it gives it its own space in the mix later without spreading it across the whole stereo field, this giving you more control for placement and suchlike later.
[Edited to add] Once you've recorded your vocal, if the waveform you're seeing onscreen looks too quiet (as in you can barely see it, aside from a few wee bumps) then don't worry. Recording with too much gain added will also raise the noise floor and potentially add problematic frequencies that'll screw with the vocal, so a quieter recording isn't necessarily a bad thing. Regardless though, I always, always normalize my dry recordings to -2dB. Most recording software will have the option to normalize, which just basically means "bringing all the levels up uniformly" - Go into one of the tabs like "Effects" or "Process" or something and look for "Normalize". Normally this'll be defaulted to 0dB, but as I've said -2dB brings it more in line with a sort of industry standard and also affords you much more freedom further down the line.
A vocal that's been recorded too quietly is easier to clean up and mix than one that's been done too loud or compressed to death. Mastering engineers HATE being sent tracks that are hitting 0.5dB or even 0dB (beyond 0dB, this is called 'clipping', which is basically distortion that really sucks and should be avoided - see picture), so I tend to mix as if it's going to be professionally mastered; even just my vocal covers using backing tracks like yours. As Bob said before, all of this is really subjective so what sounds good to one person may sound awful to another, however there are certain fundamentals that all engineers are taught or at least learn as they go so don't overcomplicate it for yourself. Stick to the basics, seriously you'll learn more by screwing up a thousand times and exploring what doesn't work than you would researching the perfect recording setup for you at the stage.
As to whether your original recording could be fixed, I agree in part with what Tony's said however, having repaired loads of really crap recordings myself, I think that it's possible to salvage a half-decent version for you with a bit of patience. I'd really need to hear the dry tracks themselves before being able to say with any certainty, but I don't think it's an entirely lost cause. To be honest, it'd be far better to go back and record from scratch and get that consistency in the performance, but if you're pressed for time then I'll happily have a play around with it and see what I can do. Just mail me the dry vocals tracks and the backing track to: theweirdnet@gmail.com and I'll see what I can do.
Something else to bear in mind is the volume at which you're playing back your recordings. If you're output volume is maxed out, everything will sound super loud and will be a false representation of the actual track. This is actually an important thing to know for mixing later down the line, especially on a home setup, like mine and yours, without proper studio monitors; I mix on various sets of earphones and small speakers just now, but the same principle applies. Try to mix while listening back a lower volume than you normally would. This encourages your ears to listen more and you can hear things that a louder playback would normally obscure. Always still play it back loud too so that you can see how it sounds, but try to stay at a lower volume until you're happy with the way the track sounds overall. Bumping up the volume later then allows you to deal with the less subtle changes that need to be made, e.g. maybe a higher vocal harmony is poking out of the mix in an unpleasant way and needs to have some EQ cut or have the levels pulled back a bit.
Hope this helps!
* The recording stuff has already been well covered.
I have an unfounded theory... that basically we become who we love... musically-speaking.
I.e. I love George Lynch's guitar playing... is it any wonder that my tone and phrasing is very reminiscent of his?
Vocally I love Eric Martin (Mr Big), Oni Logan (Lynch Mob) and Ray Gillen (Badlands)
Is it any wonder that, while I don't have the same mature distortion or top end as them (yet), my voice bears more than a passing resemblance to them when I sing pieces of theirs that I love?
Your performance just adds credence to the theory
Haha, nice theory. Well, I always listen to Freddie's vocal tracks seperately and I observe what he's doing to make the sound. Then I practice with trial and error and try to get the same sound. Of course I don't want to be a Freddie Mercury ''impersonator'' I just want to have his "vocal style" while singing like myself.
But anyways I'm glad I could contribute to your theory haha.
-Diego
All the best,
-Diego
Peace
Here we go @videoace @highmtn @Furious_Phil
Here's the master track, thanks to the help of @TommyM !
Recording is a lot of trial and error starting out. There are certain principals you can look up online that will make your recordings sound better that are fairly easy to implement.
Peace, Tony
The cool thing about really learning to record better demos is that you teach yourself to sing better along the way, by digging down deep into the song and doing some intense listening and correcting.
Nice job, Tommy and Diego.
GENERAL COMMENTS
- Due to the sort of patchwork recording, and the fact that trained ears will pick up more easily on the cuts and inconsistencies, it's hard to hear it as a continuous performance. This is really just a technical thing that comes down to the basic recording, but I can hear when you've done stuff like, for example, the "somebody too..." bit and patched in the last "looooveee" bit. There's a very clear change in vocal tone and volume compared to what came before, so I tried to balance it a wee bit and make it sound less artificial. Again, it's just a technical thing that comes down to the way you're recording and the way you're hitting the mic from different positions and proximities, but I'll offer a few tips on that later.
- In my opinion, you've nailed adapting Freddie's style to your own voice for the most part. You don't sound exactly LIKE him but his influence is clear, and you perform this song similar to how he would but with more rasp and grit. My only real criticism would be that you sound, at times, like you're still trying to sound like him and it's pushing your voice into spaces that, to my ear, sound unnatural for your overall sound. For example, and I mention this as I'm still doing the same at times, you're pushing too much pressure and not opening up the vowels on some of the high chest notes so it's sounding stressed compared to the more effortless phrases you're pulling off.
- Pitch-wise, you sound really good. I ran your vocal through a pitch analyzer and there aren't a lot of problem that couldn't be sorted with better support and compression. At times, you seem to be going a little sharp too when you're belting but it's most likely just down to going at it too hard.
- Dynamically, you really convey the emotion and ride the quieter parts fairly consistently, so there's no excessive leaps in volume between parts. That, to me, is one of the harder things to do with songs like this; there's so many peaks and troughs involved that it'd be easy to veer between super-quiet and loudly brash, but you appear to have it under control.
RECORDING CRITICISMS
(Don't take any of these comments as insults or personal criticisms; I'm just trying to point out where a few technical problems appear and I'll try to help you avoid them in future if possible.)
- It sounds like you've used a noise reduction effect. These, when used incorrectly, tend to chop off higher frequencies and deaden the sound. Don't get me wrong, they're tough to get to grips with and I know you're only just starting out so don't beat yourself up over it. Noise reduction is great and sometimes even essential, but it's something that needs to be done right to make sure your recording maintains its quality while removing the hissing and noise. Most noise reduction effects include the ability to sample the overall noise profile of the recording, which basically means that it analyses the whole track and, based on certain parameters, isolates all of the noise you don't want. The best way to do this is to start it prior to when the voice or other instrument kicks in, that way the effect samples the overall room noise, the ground noise and any other non-musical interference. From there, it's usually a matter of adjusting the parameters on the effect to reduce the noise to an acceptable level, but the specifics of this can be complex.
- Did you use any sort of pitch correction software? There are wee bits that sound unnatural in a way that, to me, sounds excessively corrected but I could be wrong.
- Your timing on the "Find. Me. Somebody to looove" bit is off. I tried to cover it by fading the backing track in a little later, but it's still there. It sounds like you're missing a few bits of harmony too, but admittedly it's a complex part and you've generally got it covered.
RECORDING TIPS
- Record your vocals in mono. Placing a stereo tracked vocal into that song was pretty difficult 'cause it's got such a wide stereo field and involves lots of panned parts, such as the harmonies and the guitars. This will make it easier for you to place your vocal against your backing track without it sounding like it's really obviously sat on top of it, if you know what I mean. I'm still very much an amateur at this stuff, even after about 15 years, so please don't take my word as gospel. I can only speak based on my experience and what I've learned, but I've definitely found that recording all your parts in mono makes mixing a whole lot easier. There are times when recording in stereo is totally appropriate, but I think for home studio folks like ourselves it's more helpful to stick to basics and do stuff that makes it easier in the long run.
- Maintain a fairly consistent proximity to the microphone. Obviously you're going to move back and forth a bit as you ride the performance, balancing your volume and all that stuff, but try to stay roughly in one place.
- Don't stand too close/too far away from your mic. Too close and your mic will unnaturally emphasise the bass frequencies in an unpleasant, problematic way; too far away and you introduce too much space and inconsistency into the sound, also it exposes more ground noise and can make for problems later in the mix. A pop shield is essential to avoid plosives and reduce the "ess" sound, but careful positioning in relation to the mic can reduce both with a bit of practice.
- Try panning your own harmony parts to the left and right; play around with it and see what sounds good. At the moment, they're all compressed together so it's hard to hear them and they're really, really good so they should be more clear. Remember, when you're mixing a track you're trying to create a three-dimensional space where all the sounds have their own place, just like your living room or bedroom. Your main vocal should be front and centre, no question. Backing vocals, depending on how many you've got and the function they're serving (e.g. harmonies, double-tracking, etc) should have their own place. I imagine it like watching a band onstage and where each sound would be coming from physically, so you've got your guitars at the sides, drums punching in the centre but the cymbals and hi-hats sizzling away up top and at the sides, your bass sitting right there on the ground, etc.
- If you're using the Yeti, keep it set to "Cardioid Mode" and keep the gain level down to a reasonable setting, i.e. not overloading and going into the red. Make sure you've got it set up facing you too, don't sing into the top of it like a dynamic mix; cardioid mics record sound coming straight at it from the front.
Hopefully this is helpful, but remember it's just my opinion and other, more experienced folks may be able to contribute more or correct my errors. If you want me to try mixing anything else for you, I'd suggest doing the following to ensure that I can make it sound as good as possible:
- Send only single, mono vocal tracks, i.e. send each vocal as a separate track, so for example: Main Vocal, Harmony 1, Harmony 2, etc etc. Only one vocal to a track, don't comp together your harmonies into one track as that means I can't split them up and spread them out so they stand out.
- Don't apply ANY effects to it at all. Just the bare bones track is perfect, warts and all.
- Send the backing track you used so I can make sure I'm sticking to the same timings as you.
All the best!
- Tommy
Oh my, how do I reply to this? LOL.
Yes, I agree. Now that I take a look at the original recording, I'm like wow. "What am I doing?"
All of your comments just opened my eyes to stuff I was blind to lol.
No, I do not use any kind of pitch correction software. I usually stay away from them, because I like a natural sounding voice.
I developed a SIMILAR style to Freddie Mercury's I kind of adapted to his way of singing. But I usually like throwing other singers like David Bowie in there like you may have seen in Under Pressure.
I don't remember I MIGHT have used noise reduction? I don't even know anymore.
I like to stand a little closer for lower notes, and when I am about to belt I usually back away from the microphone so It doesn't distort or anything. So what exact distance should I be to make both sound good without having to move?
Oh believe me, I started panning my harmonies already.
I usually don't apply that many effects, the one I have used for other songs, not this one is reverb (I think "Take On Me" only)
-Diego
I see, I'll be pretty sure to be more careful next time! Haha.
Exactly it's really cool to learn new stuff all the time. As they say, there is always room for improvement. So here I am.
Thank you for the feedback! :-)
-Diego
It's funny listening back to your own recordings with hindsight. I've got the stuff I did when I first started out (just weird mini-sound scenes using only voice) and it's hilarious, but a great reminder of how and why we continue to learn and develop. It'll all come with time, just have fun with it and you'll find yourself able to produce a really high-quality product before you know it.
When it comes to pitch correction, a lot of people are quick to criticize it. From a production perspective, being able to correct small errors in a performance rather than re-record the whole thing can be a massive time saver. Done properly, pitch correction should be completely transparent and adjustments of tiny percentages can make a huge difference, especially with harmonies. Obviously we should be aiming to sing without having to use any sort of corrective software or hardware, but when it comes to later in the mixing stages it can be a lifesaver, e.g. I've done vocals that were perfect sounding in terms of the performance but the tuning was flat in places. I'd already packed away all my gear and it was later at night when I was trying to mix it so there was no way of going back and re-recording a better take. I was able to pull the flat notes up by the 1/8th of a note they'd fallen short (which is negligible alone, but as part of a harmony was noticeable) and avoid a complete re-recording as well as all the variables that entails.
All in all though, they're best avoided unless you know what you're doing. Again, this is something I've spectacularly screwed up, and also had screwed up for me by a professional engineer, so don't make the same mistakes I did...hahaha!
With regards to mic proximity: Test for yourself to see what gives you the best sound as it's a combination of factors. The space you're in will colour the sound, so a bigger, more spacious room will add a natural reverb that'll interfere with any post-recording effects you're applying. (For example, on your "Somebody to Love" my ear can hear the room you're in, especially on the higher, more reverberant notes) Conversely, a small, acoustically deadened room is ideal as the mic will capture a straight recording with minimal environmental distortions to the sound.
The distance will vary depending on lots of factors. There's video of Michael Jackson recording a vocal with Quincy Jones in a huge space and he's standing a good 5-6 feet away from the mic. There's also footage of Freddie Mercury doing the vocal for "One Vision" where he's standing at, what I would consider to be a more 'normal' distance from the mic.
Personally, based on hearing your basic recording and taking everything into account, I'd suggest standing maybe a foot away from the mic but continue to do what you're doing in terms of moving your head closer/further away depending on volume, etc. Try to keep your body stationary if possible, that'll avoid rumbles and the noise of the mic stand or surface area moving. These are things that can be cut out later in the mix, but avoiding them entirely will make life SO much easier for you.
With panning, there's a few things to be careful of: Watch out that your parts don't 'poke' through the mix in an unnatural sounding way. If you pan parts to hard right and hard left, there's a risk that they'll sound as though they're sitting 'outside' of the track. If this happens, bring them in to around 75-80% and it should sound a little more natural. Hard panning is really, really useful at times and particularly on instruments like guitars so that they can be separated in a mix, but with only vocal parts it's worthwhile to be cautious and pay attention to the whole sonic picture. Also, and this is really more useful if you're doing a bigger mix so it may not be relevant but I'll mention it anyway, it's good to use slightly different EQ setting on the panned vocal parts. That way, you can preserve the character of each part, but place them in the mix without their frequencies clashing and causing problems.
Effects are a complex area and can really make or break your recording, especially if applied without due care. Reverb is a great example of a commonly used effect...that many people completely wreck their recordings with. Reverb creates an artificial space for your sounds to sit in, partly to 'glue' them together in one acoustic space and also to give depth, sustain and colour to a sound. The problem is that reverb can also introduce unwanted frequencies into your sound and cause it to be really muddy and dull, unless it's done properly.
Another factor with effects is the way in which they're applied. In most basic software like Audacity, you can only apply the effect once to the whole, for example, main vocal. Once it's applied, you can't go back and adjust the parameters to further shape the sound to fit your needs. This is why a half-decent, even free DAW (Digital Audio Workstation) like Reaper can offer so much more flexibility: You can use the effects as what's known as an "insert", i.e. you can insert it into a chain of effects that will affect the sound AFTER it's been recorded. That also allows you to adjust settings and all that sorta thing so that you can find the 'right' sound for you.
Better still, with a DAW, you can set up what are known as "Sends" or "Returns": You can set up an effect like a reverb and then send the signal from, for example, your vocal to the effect and then control how much of the effect is blended into the whole track. This is MUCH more flexible and ideally what you want to aim for as it'll open up so many new possibilities and avoid muddyness or overpowering effects.
If you're just using something like Audacity, one workaround to get a similar blended effect without sends is to duplicate your dry vocal and then apply the effect to the copy. Bring both volumes right down and then gradually raise them until it's at a level you're happy with. Ideally, you'd want to apply EQ and cut all frequencies below, for example 150Hz, on the reverb track to avoid cluttering but that's really just an added touch.
As I'd said before, if you want me to mix down one of your tracks from the pure dry vocal parts, I'll gladly do so, just to give you an example of what's possible with a basic setup. All of this stuff will fall into place soon enough, but in the meantime just keep educating yourself and asking questions. I'm more than happy to offer what little I can, so please don't feel bad about asking more on this sorta thing; I love talking about it! Hahaha!
With, for example, Audacity, as we've seen we need to apply our effect to the whole track as a one-shot process. We can't go back and edit it without first undoing the effect and then re-applying it with different settings, so it's a pain in the hole and limits what we can do. In a DAW, in my case Cubase, once the dry audio is recorded we can add "Insert" effects to the channel, as we've seen, which then alter the sound AFTER it's been recorded. Furthermore, we can also adjust it while the audio plays, only finalizing the sound when we export the track either by itself or as part of the whole mix. I could go further into this, but I just want to give you an overview based on a few bits and pieces I've learned. Anyhow...
With "Insert" effects, it's important to know that the order in which you add them can and does change the dry sound in different, sometimes undesirable ways. That sounds like common sense, and it is, but there are technical reasons too. The same applies when doing the one-shot effects in Audacity or whatever; the order you apply the effects can be optimised to produce better output quality.
So, without overcomplicating things any further, what I'm trying to say is this: Your effects chain, in my opinion and based on what works for me, should be similar to the following (let's use a vocal track as an example):
1. Normalize your dry track to -2dB
2. Use EQ to cut everything below at LEAST 90Hz. Ideally use something like Voxengo SPAN to see exactly where your vocal take sits on the frequency range, but from what I've heard from you I think you're safe to go as high as around 100kHz without affecting your sound.
3. Compress the track gently, so use an attack speed of around 300-500ms and either use "Auto-Release" (if it's available) or set a faster release rate. Use a ratio of around 2:1 and experiment with the threshold until you can hear it just catch the louder transients. All you want to do is to balance out the sound overall and give it consistency, rather than being super quiet and then super loud.
4. At this stage, you'll have a reasonably clean track that you can then work on further. After this, I'd normally do some more compression to bring everything up and then EQ the track to enhance the higher end (that 'sparkle' and brightness), boost around 2-4kHz, cut out a few dB of gain from around 150-200Hz to reduce the boxyness and cut out any remaining low-end rumble that's been added with the compression.
The order in which you place your effects can change things dramatically. Things like compression and limiting will also raise your existing noise floor, as well as boosting noise in the higher 14-18kHz range so be aware of this. It can all mainly be fixed so don't worry, but this is why it's important to ensure you're getting a good, strong and clean signal going in from the beginning. You know the old saying: "You can't polish a turd"...or, more politely, "garbage in, garbage out".