Propellerhead Forum

Propellerhead Forum (https://www.propellerheads.se/forum/index.php)
-   General Forum (read only) (https://www.propellerheads.se/forum/forumdisplay.php?f=3)
-   -   Patch help request: Crazy Dubstep bass that sounds like a voice (https://www.propellerheads.se/forum/showthread.php?t=135112)

AndrewBest 2011-01-06 00:04

Patch help request: Crazy Dubstep bass that sounds like a voice
 
Listening is easier than me explaining. Skrillex uses the sound at 1 min into this track: http://www.youtube.com/watch?v=WSeNSzJ2-Jw. And Feed Me uses it at 4 min into this track http://www.youtube.com/watch?v=Uk9nl3j1VDk. Can anyone help me turn a standard dubstep bass into this? I get the basic part about using the LFO to modulate the filter frequency. And typically using a square wave oscillator, but this is more complicated. Any suggestions on getting the vocally sound? I tried a formant filter but it doesn't sound right. Thanks.

3rdFloorSound 2011-01-06 00:15

Sounds like it's vocoded to me.

Droog 2011-01-06 02:28

abletons vocoder just a guess. reaktor lazer base is a very good tool for dub step as well.

this is pretty shit music tho the way it goes from trancy bollocks to "pfuhprfuhwepfhswfpu" vocoder base.

James Bernard 2011-01-06 03:19

The Formant filters on Thor will get you this sound.. With distortion added somewhere in the effect chain of course.

Bretstradamus 2011-01-06 05:02

Quote:

Originally Posted by AndrewBest (Post 830450)
Listening is easier than me explaining. Skrillex uses the sound at 1 min into this track: http://www.youtube.com/watch?v=WSeNSzJ2-Jw. And Feed Me uses it at 4 min into this track http://www.youtube.com/watch?v=Uk9nl3j1VDk. Can anyone help me turn a standard dubstep bass into this? I get the basic part about using the LFO to modulate the filter frequency. And typically using a square wave oscillator, but this is more complicated. Any suggestions on getting the vocally sound? I tried a formant filter but it doesn't sound right. Thanks.

YEEEEEEAAAAYYUUUHHHHHHHH SKRILLEX. Here's my quick and dirty tutorial:

If you've snooped around on any dubstep forums, you'll see his name more than a few times and you'll see posts just like yours all over. Most of people around those forums use Massive, but luckily you've got Reason so you can do everything they do much easier and there's many things you've got that they don't.

Start with the bass. You want talking bass? James Bernard was right with the formant filter, but that usually comes later and there's a reason that yours doesn't sound right. It really takes a lot of work from the very beginning to get the right kind of sound for the talking bass, so that's why I'm going to start from the very beginning.

First, you need just a great bass sound to start with. To get basses that sound like Skrillex's, you'll want to look into detuned saws or PWM'd squares. Saws and squares are great starting points, saws are more common. The key is to get movement in your sound. Pull up a Thor and throw in three sawtooths and detune them. You can do this however you want, but typically I have the first two detuned the same amount in opposite directions and the third one is in the middle, they tend to be cleaner this way. For more movement, I like to make this a true stereo patch (as described on the back of Thor) and then have two modulated notches in Filters 1 and 2 and give them different rates of movement. Generally, keep the rates slowish. Sometimes a comb filter can work in the third filter spot, if you want that kind of sound. If you're using wavetables as oscillators, you can modulate the crap out of them with the position knob, but of course there's an endless possibility of ways to modulate something in Thor I don't think I could describe them all!

Next, once you're done playing around with oscillators and filters in Thor you may want to do some more processing with Chorus, Flanger, short Delays, more filters, or any kind of Scream 4 distortion. A couple techniques you can use while doing these are frequency splitting and resampling. Frequency splitting is separating the sound with the MClass stereo imagers into two or more bands. I typically do three: low, mid, and high. On the lows you don't want too much stuff going on because that can make the sound a bit unstable, I generally use tape saturation or tube distortion, light choruses or nothing at all.

OH one more thing, you may want to make a separate synth with a sine oscillator for the sub (if you want a sub shaking synth). When it comes time to wobble, a low pass won't do anything to it, you'll need to modulate the amp volume instead. You don't need to do any effects on the sub, it won't sound like much anyways. Hopefully you've got good speakers if you want the sub.

On the mid range frequency I'll usually add anything with slight movement or different kinds of distortions. When processing mids and highs you'll want to use light amounts of everything or it'll end up sounding all messy. It's kind of like cooking, get just the right amount but not too much.

On the highs I go crazy with these. Phasers, choruses, distortions, anything at all. Don't be afraid to use a little faster rates on these things here, but remember not too fast.

Don't forget that when you've merged these frequencies you can also use all these effects as send effects, too.

One thing I forget to mention so far is BIT CRUSHING!!!!!!!!!!!!!!!!!!! That helps the sound a lot, but remember with all the effects here you'll want to use light amounts of it.

The other technique I mentioned is resampling. That's where you take a sample of you sound (like just a long note) and sample it into a sampler. Why would you want to do this? Several reasons.
1. You can save on CPU resources, especially if you like using lots and lots of effects. Reason is much more CPU friendlier than other software, so this isn't something we always needs to worry about.
2. You can do the same effect more than once!
3. If you end up with a mess you can go back to an earlier version and start again from there. This is where resampling is most helpful, I think.
Resampling in Reason is much easier than any other software, thanks to the recent upgrades. Just sample your sound into an NNXT.

So this is a really long post (sorry), though I'm trying to make it concise for you here. You don't have to do any frequency splitting, resampling, or any other effects if you don't want but they are just different ways to get different sounds. There's a lot more to all of these steps, but I haven't even gotten to your original question yet, the talking sound! It really takes a lot of work from the very beginning to get the right kind of sound for the talking bass, so that's why I started from the very beginning.

The talking sound is created using modulated filters, or for us it much easier with the formant filter in Thor. Route your polished bass sound into a Thor and pull up that beautiful red module. This thing doesn't get enough credit. I highly recommend you read this article: http://www.propellerheads.se/substan...article=part38
At the end it'll teach you how to make the sound say "WOW". Since I think you should read it, I'm not going to explain it here. Go read it! Now! There's many more ways to make your bass say things, like "YEAH" or "YO" or "WAAAHH" or "YAAAHH" or anything else. The trick is finding how to make it say these things. Don't forget about the voice oscillators in Thor's wavetables or in the Malstrom!

Does all this answer your question? I typed it really fast and tried to make it concise but unfortunately this may leave you with more questions than answers. Let me know if you need me to clear any vague information up for you or if you'd like more detail about something. These kinds of sound take time so sit back, relax, and have fun!

Anomecron 2011-01-06 05:15

Quote:

Originally Posted by Droog (Post 830481)
abletons vocoder just a guess. reaktor lazer base is a very good tool for dub step as well.

this is pretty shit music tho the way it goes from trancy bollocks to "pfuhprfuhwepfhswfpu" vocoder base.


It is a tad ummmm "Gorgonzola" I must confess.
The bass sound itself is rather good though.

Droog 2011-01-06 05:42

i do agree this is kinda the idea James to make the A E I O U sound.

do you think though for these sort of sounds, magnus left users better goodies in the malstrom?

wavetables =

ELEKTRONIK

FORMAT SAW

THROAT

im sure there are others which make that aaaaaaaaaaaaaaaaaaaaaeeeeeeeeeeeeeeeeeeeeeeeeeiiii iiiiiiiiiiIKnowReasonsTeamWillUpgradeuuuuuuuuuuuuu uuuuuuuuuOOOOOOOOOOOOOOOOOOMalstrommmmmmmmmmmmmmmm mmmmmmmm

:D

Bretstradamus 2011-01-06 06:17

Quote:

Originally Posted by Droog (Post 830481)
abletons vocoder just a guess. reaktor lazer base is a very good tool for dub step as well.

this is pretty shit music tho the way it goes from trancy bollocks to "pfuhprfuhwepfhswfpu" vocoder base.

BOOOO don't be mean! You can use a vocoder as well but it's not always necessary. And skrillex doesn't use lazer bass, to my knowledge (but don't quote me!).

AndrewBest 2011-01-06 08:48

Thank you for the very thorough response. It gives me plenty of things to try out. I'll follow it step by step and I'm excited to hear the results. With a very quick and dirty test of the vocoder method I find that it sounds better than my simple formant method. And by better I only mean more like a replication of what they do in those tracks. But reading the article you mention could definitely change my opinion of the formant filter after I can use it better. Thanks again for the tutorial. I'll check back in after following it.

picass069 2011-01-06 15:48

not my tutorial, but this might be what you're looking for:

How to make Skrillex's signature talking bassline synth

http://www.youtube.com/watch?v=pOgo7RXla5g


All times are GMT +2. The time now is 10:21.