Auto Leveler, Multi-Band Compressor, and Frequency Split in Bitwig
Tutorial | Feb 13, 2024
In this video, I cover three different topics related to EQ techniques. The first is an auto leveler method using bandpass filters and an amplifier to maintain a consistent volume. The second is a multi-band EQ compressor, created by combining an EQ with a grid compressor, allowing for individual frequency control. Lastly, I demonstrate the use of the Loud Split device to selectively affect different frequency ranges in a sound, helpful for removing noise or enhancing specific elements.
You can watch the Video on Youtube - support me on Patreon
In this tutorial, I explored several innovative techniques to achieve auto EQ effects similar to the Waves Factory Equalizer within a digital audio workstation using stock devices, primarily focusing on Bitwig Studio. Here's a summary of the methods I discussed:
-
Auto Level Preset Technique: Utilizing my previously designed auto level preset, which functions as an auto leveler by dividing the signal by the signal from a follower, achieving a consistent output level. This technique was adapted to create a multi-band auto leveler using an XP filter and bandpass filters, with voice stacking for frequency selection and modulation for frequency distribution adjustments.
-
Multi-band EQ Compressor: By leveraging the EQ+ and FX grid in Bitwig, I constructed a multi-band EQ compressor. This setup includes bandpass filters for frequency analysis, feeding into a grid compressor for dynamic control across multiple frequency bands. This approach allows for detailed frequency compression, with the capability to adjust the threshold, ratio, and response times for each band.
-
Loud Split Device Utilization: Demonstrating the use of Bitwig's Loud Split device, I showed how to separate audio into loud, mid-loud, and quiet parts. This method enables targeted processing of different parts of the audio spectrum, such as reducing harsh frequencies or enhancing quieter elements, with adjustments for tilt to emulate pink noise distribution.
-
Frequency Split and Limiting: Lastly, I discussed using frequency splitting to divide the audio spectrum for targeted limiting, achieving a spectral limiting effect. This can create unique sound alterations but is noted for its potential harshness if overapplied.
Each technique was crafted with creativity and practical experimentation in mind, aiming to achieve professional audio processing results without relying on specialized plugins. While acknowledging the limitations and potential phase issues with these DIY approaches, I emphasized their utility for sound design and mixing within Bitwig Studio.
Additionally, I mentioned that while these techniques offer great flexibility and creative potential, dedicated plugins might provide more precise control and quality for specific tasks. Nonetheless, these methods serve as valuable tools for producers looking to explore sound design and mixing possibilities within their DAW using stock devices.
For those interested in trying out these techniques, I've made the presets available on my Patreon page, encouraging experimentation and adaptation to personal production workflows.
Questions & Answers #
Maybe you dont watch the video, here are some important takeaways:
How can Waves Factory Equalizer be created with this technique? #
To create a Waves Factory Equalizer using this technique, I used an auto level preset and applied it to a bandpass filter. This allowed me to create a multi-band auto leveler that can be used for sound designing and EQ purposes. While it may not be as precise as a dedicated plugin, it can still provide rough frequency balancing.
Can a multi-band EQ compressor be created using the EQ Plus and FX Grid? #
Yes, a multi-band EQ compressor can be created using the EQ Plus and FX Grid. By applying Zen key filters and stacking analyzers, multiple frequency bands can be analyzed and sent through the grid compressor, allowing for individual compression and frequency control. While it may not be of the highest quality, it still serves as a fun and functional tool.
How can the Loud Split device be used for EQ purposes? #
The Loud Split device can be used for EQ purposes by selectively disabling or reducing the volume of certain frequency bands based on their loudness levels. By utilizing the tilt function, you can adjust the frequency distribution to match a desired curve. This can help clean up the spectrum and make sounds clearer, especially for removing noise in recordings or toning down overtones.
What is the overall usefulness of these techniques in Bitwig? #
While these techniques may not provide the most precise and professional results, they can still be useful for sound designing, rough frequency balancing, and creative EQ applications. It's important to remember that these techniques are not meant to replace dedicated plugins, but rather provide alternative options for experimentation and exploration within Bitwig Studio.
Transcription #
This is what im talking about in this video. The text is transcribed by AI, so it might not be perfect. If you find any mistakes, please let me know.
You can also click on the timestamps to jump to the right part of the video, which should be helpful.
[00:00:00] So in this video, I want to cover a lot of different topics.
[00:00:03] Please stay with me, stick with me.
[00:00:06] I try to make it short and straight to the point as possible.
[00:00:10] So on my last video, LA Winter asked,
[00:00:12] would it be possible to make something like Waves Factory Equalizer
[00:00:16] with this type of technique?
[00:00:17] You know, I showed like the EQ and how you can apply
[00:00:21] different LFO curves to that and so on.
[00:00:23] And the equalizer here is basically some kind of auto EQ
[00:00:27] where you can apply pink noise or 4.5 dB per octave noise
[00:00:31] to audio material in real time all the time.
[00:00:35] So this is probably made with FFT,
[00:00:39] splitting up the audio signal into multiple frequency bins
[00:00:43] and then applying different audio normalization to that, I guess.
[00:00:47] So this is probably hard to do inside of a door
[00:00:51] just with stock devices, but I tried it anyway, just for you.
[00:00:56] So my first idea to do this was to actually use my auto level preset.
[00:01:04] Auto level.
[00:01:06] Looks like this here.
[00:01:07] I made a video about this, I think two years ago,
[00:01:12] just dividing here the signal by the signal from the follower,
[00:01:17] which gives you kind of just an auto leveler.
[00:01:20] And to show you how this works, I put here a peak limiter at the end
[00:01:25] and in the front I adjust the tool device
[00:01:28] and we decrease here the fall time so the auto leveler
[00:01:31] reacts more quickly to changes.
[00:01:34] So when I hit play, you can see the peak limiter basically
[00:01:38] scratches here 0 dB line.
[00:01:40] So this auto leveler tries to amplify the signal
[00:01:44] so the highest peak of the signal goes to 0 dB at all times.
[00:01:49] And if it's lower, it tries to increase the signal.
[00:01:52] So it looks like this.
[00:01:54] [Music]
[00:01:56] You can see 0 dB, right?
[00:01:58] Now I decrease and increase the signal before the auto leveler.
[00:02:02] [Music]
[00:02:05] You can see it stays at 0 dB.
[00:02:06] [Music]
[00:02:09] Stays at 0 dB.
[00:02:11] So the volume changes here are basically inverted by the auto leveler
[00:02:18] instantaneously because I also decreased here the fall time.
[00:02:22] So the auto leveler reacts basically like an inverted compressor.
[00:02:26] Instead of pushing signals down and tries to
[00:02:30] amplify the signal until it reaches 0 dB.
[00:02:33] So I really like this and I use this a lot sometimes
[00:02:36] just for sound designing when you have a lot of EQ things,
[00:02:40] right? You EQ a lot of things here and there
[00:02:42] and then the volume changes all the time.
[00:02:44] And then to put the auto leveler on there and you know
[00:02:48] the volume is basically taken care of.
[00:02:51] So instead of using this auto leveler in full range,
[00:02:56] my idea was to actually make this multi-band auto leveler.
[00:03:02] So we just use an XP filter,
[00:03:05] pull the key tracking down and we use here a pretty wide bandpass 2 pole filter
[00:03:13] and also we pull here the resonance down.
[00:03:16] And everything here goes into the bandpass filter of course.
[00:03:21] Also the output, the audio output here, it's important.
[00:03:24] And then we use a value in bipolar mode and we do basically the same thing we did
[00:03:31] a few videos ago with the stacking.
[00:03:35] So here I want to also use a quantizer, which is nice.
[00:03:38] So we want to stick here to certain frequencies, right?
[00:03:42] We want to stick to a scale.
[00:03:43] So all the frequency we choose land basically on the scale of D# minor now
[00:03:48] because this track is made in D# minor.
[00:03:51] So it's a musical, more like a musical approach to that.
[00:03:55] Okay, so then we want to use an amplifier here at the end
[00:04:01] because it gets loud pretty quickly I guess.
[00:04:06] Yeah, let's put it here.
[00:04:10] And then we want to use voice stacking 16 voices.
[00:04:15] So we have 16 bandpass filters now.
[00:04:17] We pull this up and use a stack spread modulator.
[00:04:21] Minus 1 to plus 1 and then we spread out to the frequencies.
[00:04:27] And to see which kind of frequencies we are selecting,
[00:04:30] I'm using an EQ plus and then we increase the resonance until we have self resonance.
[00:04:44] So you can see here the frequency positions.
[00:04:48] Maybe push this here too very fast and huge.
[00:04:53] So we can now change this a bit.
[00:04:59] We probably also want to have the center frequency around 1k.
[00:05:08] So I go for C5 here.
[00:05:12] And then...
[00:05:16] Maybe we'll be offset this here.
[00:05:22] So this is here the highest peak.
[00:05:29] Then we have a lot of low peaks here.
[00:05:31] Maybe we don't need that much filters here.
[00:05:36] Maybe 13 is enough.
[00:05:38] So now we can also, instead of using this amplifier, we're using a different amplifier.
[00:05:48] We can say this changes basically the frequency.
[00:05:53] So the higher we go with the frequency, the lower we want to have the volume.
[00:05:58] So we turn the volume down.
[00:06:06] So we can tilt the whole thing in terms of frequency distribution.
[00:06:12] I go to minus 1.
[00:06:13] And then I use here another macro.
[00:06:16] And I modulate here the amplifier with that.
[00:06:21] So now we can call this tilt.
[00:06:26] And we pull this down.
[00:06:30] Okay, so I guess this should be already working.
[00:06:36] So let's try this out.
[00:06:38] [Music]
[00:06:40] [Music]
[00:07:08] Maybe play around here with the distribution of this, of the frequencies.
[00:07:20] With higher resonance you can hear or you can already also see on the EQ where the
[00:07:35] frequency spots are.
[00:07:37] [Music]
[00:07:54] So this sounds, this is how it sounds without the auto leveler here.
[00:07:59] [Music]
[00:08:17] So with the tilt you can tilt it more into the warm spectrum or like you probably go up to
[00:08:23] 60 pp/oct here, I don't know.
[00:08:26] And this is probably pink noise.
[00:08:27] [Music]
[00:08:35] So everything is not precise.
[00:08:36] It's more like tuned to my ear and you know roughly designed kind of.
[00:08:42] But I kind of like this approach because you always need to hear and you know check with
[00:08:48] your ears if it sounds actually right or wrong.
[00:08:51] [Music]
[00:09:02] So this is basically just an auto EQ based around this auto leveler and bandpass filters
[00:09:10] here or pretty wide bandpass filters.
[00:09:12] And it kind of works, kind of works.
[00:09:15] You probably run into problems when you use this multiple times in a row because then you
[00:09:20] have some phase issues, right?
[00:09:21] You have overlapping frequencies when the bandpass filters, but you can try and play around
[00:09:26] here with different voice stacking numbers.
[00:09:30] Maybe 13 is too much.
[00:09:32] You can also try it out with 8 or just 4 or something like this.
[00:09:36] So it's all you know try and error.
[00:09:40] So this is my first idea for that.
[00:09:42] I put you this preset also onto my Patreon.
[00:09:48] My second idea was to actually create some kind of multi-band EQ compressor.
[00:09:58] So based on my yesterday's video, I just took here the EQ plus
[00:10:02] and strapped it onto an FX grid again.
[00:10:06] And then in here I built some kind of stacked analyzer.
[00:10:12] So we have bandpass filters in here analyzing multiple frequencies
[00:10:16] and then using these frequencies and sending it through a compressor, a grid compressor
[00:10:22] made by the famous class who is a Bitwig developer and also inventor of the OTT preset.
[00:10:29] So it's an official confirmed good quality compressor inside of the grid.
[00:10:37] He did this as a preset for when the grid released three years ago.
[00:10:44] So I took this compressor basically and made it multi-band.
[00:10:47] So I used a lot of Zen key filters here also pretty wide,
[00:10:51] BP2 filters to analyze frequencies.
[00:10:54] So we have multiple bands here, eight bands.
[00:10:56] One band is actually at the moment active.
[00:10:59] Let's turn this down a bit.
[00:11:07] You can see this frequency here is reduced.
[00:11:11] You can change the frequency.
[00:11:13] We can change the threshold.
[00:11:14] We can change the ratio.
[00:11:15] How much this is actually compressed.
[00:11:17] We can change the fall time and the rise time of the follower.
[00:11:22] It's more like a second decay basically.
[00:11:25] We can change the queue here.
[00:11:30] And you not only change the queue of this,
[00:11:36] the queue of this EQ parameter here.
[00:11:39] You also change the queue of the Zen key filter.
[00:11:42] I added here this modulation where you go from BP2 up to BP8.
[00:11:47] So it gets steeper.
[00:11:48] The steeper the parameter goes here.
[00:11:52] You can also add another one band to threshold ratio up.
[00:12:07] Just reduce some high frequencies here.
[00:12:23] Okay.
[00:12:31] So you can do this with eight bands up to eight bands.
[00:12:33] You can also open this here if you want to.
[00:12:37] Then there's your global tab where you can change all the queue parameters,
[00:12:43] the quality parameters.
[00:12:45] You can change the global rise and fall time.
[00:12:48] You can also add here a look ahead.
[00:12:51] So the audio signal is just delayed by a small amount.
[00:12:54] So the processing can catch up.
[00:12:57] You can also use a side chain if you want to push this on here.
[00:13:00] And then you select basically here a different channel.
[00:13:04] Let's say here drums, top loop, whatever.
[00:13:07] And then it uses this side chain to analyze the signal and then apply
[00:13:13] all the gain reduction to your current signal.
[00:13:16] So it takes here the hi-hats or the rides or the hi-hats.
[00:13:22] I think it's just hi-hats.
[00:13:23] And reduces basically the frequencies on the master.
[00:13:27] You can also switch this off.
[00:13:32] So now it takes basically the signal here from this channel.
[00:13:35] So it's a multi-band compressor based on an EQ.
[00:13:39] And it's also, like I said, it's not precise.
[00:13:42] It's probably not super good quality, but it kind of works.
[00:13:49] This is always the case with these kind of devices or presets indoors.
[00:13:54] It's not always precise, but it's fun to do.
[00:13:58] And it kind of works roughly, but it's always better to use a plug-in,
[00:14:03] you know, where the interface is exactly made for this kind of device.
[00:14:09] And it's highly or it's more precise in terms of frequency splitting and applying signals.
[00:14:15] So this is my second idea here.
[00:14:19] I also put this on the Patreon post if you want to try this out.
[00:14:22] And then the third idea is also something I already showed you.
[00:14:28] Back when the spectral devices came out, we have this device called loud split, right?
[00:14:35] And the loud split device is basically like a Sooth or DSEQ3.
[00:14:40] It splits the band into loud parts, mid-loud parts and quiet parts.
[00:14:47] So it sounds like this and we can just disable here all the quiet parts or the mid- quiet parts
[00:14:56] or the mid-loud parts.
[00:14:57] And only stick with the sounds that basically exceeds this threshold here, this minus 29 dB.
[00:15:10] It's probably not the right dB.
[00:15:11] And then we can apply here tilt, right?
[00:15:15] 3 dB is pink noise, basically 4 dB, 4.5 dB.
[00:15:22] So we tilt basically the analyzing part and everything that goes above this line here
[00:15:27] is in the stop part.
[00:15:29] So in this case, it's only the high frequencies here.
[00:15:38] And all we have to do is we can tone this down volume wise here, right?
[00:15:43] Maybe just...
[00:15:48] [Music]
[00:16:02] You probably don't want to disable this here because it completely removes all the bins,
[00:16:07] all the frequency bins that go above this line.
[00:16:10] So you want to reduce this only in volume or you want to add here maybe a compressor or limiter.
[00:16:16] At the top part here, right?
[00:16:19] To just reduce the loudest parts.
[00:16:22] So this is also nice for the de-essing and getting rid of some harsh frequencies at the top.
[00:16:28] You can also do something stupid like increasing the loudness or the volume of quiet parts.
[00:16:37] [Music]
[00:16:44] Or completely remove here these kind of noise signals by just disabling this.
[00:16:50] With this, you can clear up basically the spectrum a bit and make sounds clearer.
[00:16:57] It's also nice for drum loops where you have a lot of noise in there.
[00:17:00] Or sometimes when I record some folie stuff with my microphone, I use this here to just
[00:17:08] remove all the noise from the microphone or some clicks and pops and so on.
[00:17:12] So this is pretty nice for that.
[00:17:14] And like I said, you have this tilt in here, which is pretty nice.
[00:17:19] And you can play around with the rise and fall, which is also only for the analyzing part here.
[00:17:26] But it gives you some bins for a lot longer in this top part here to react to it with compressors or limiters.
[00:17:36] So this is also a way to EQ stuff even though it's not made for EQing.
[00:17:42] Then we have here the frequency split, which is also nice.
[00:17:45] And by the way, I made a complete video about this here.
[00:17:48] I link also this video in the description below.
[00:17:50] Then we have here the frequency split, which splits up the spectrum into multiple parts here for boxes.
[00:18:00] And then there's a limiter at the end, so we can pull this down.
[00:18:05] [Music]
[00:18:12] [Music]
[00:18:14] So it limits everything spectrally.
[00:18:17] It sounds pretty weird if you overdo this, but sometimes
[00:18:20] [Music]
[00:18:22] you pull this down and this is here probably a white noise, a white noise curve.
[00:18:28] But if you do this just to the overtones here, you can reduce sometimes or limit sometimes
[00:18:36] some overtones, but I never use this actually.
[00:18:39] For me, it's quite too harsh.
[00:18:41] [Music]
[00:18:50] You can also, if you split this here into four parts, you can put an auto leveler into that.
[00:18:55] That's also possible.
[00:18:57] Let's open this up here, put this there, there.
[00:19:07] [Music]
[00:19:13] And there.
[00:19:13] [Music]
[00:19:24] And then because everything is the same loud as our peaks at 0 dB, you have to also put the
[00:19:28] auto leveler in the post FX here to bring this actually back to 0 dB.
[00:19:32] [Music]
[00:19:40] So with this, you can also basically auto EQ stuff here.
[00:19:43] It's not like 16 bandpass filters like I showed you before.
[00:19:47] It's more like only four boxes.
[00:19:49] But in my opinion, most of the times when you put this on a master, you only want to have a
[00:19:55] rough matching with something like pink noise or 4.5 dB octave curve.
[00:20:01] You want to have only a rough match.
[00:20:03] You don't want to remove all kinds of small little resonances because this is basically
[00:20:11] the sound itself.
[00:20:12] You don't want to remove the character of the sound.
[00:20:15] You just want to balance roughly the frequency distribution.
[00:20:20] So these are my ideas for this type of stuff.
[00:20:24] Video took 20 minutes.
[00:20:26] Okay.
[00:20:28] So all of that stuff is in the description below.
[00:20:31] Most of the presets are on my Patreon.
[00:20:34] Thanks for watching and also thanks to Virtual Riot for bringing all these topics up and making
[00:20:41] these topics popular or famous.
[00:20:44] Because I think these are quite interesting to do inside of Bitwig, Ableton and probably also
[00:20:53] FHR Studio.
[00:20:55] Sorry, it's not FHR.
[00:20:57] Sorry.
[00:20:59] Oh God.
[00:21:00] Whatever I've done.
[00:21:01] Thanks for watching.
[00:21:03] Leave a like and subscribe to the channel.
[00:21:05] Thanks for watching and see you in the next video.
[00:21:07] Bye.