How to Convert Your Microphone into MIDI Notes in Bitwig Studio for Beatboxing and Triggering Samples
Tutorial | Jun 05, 2023
In this video, I show how to record your microphone and convert it into MIDI notes to trigger random samples in Bitwig Studio. To do this, I use a trigger to trigger the kick drum, snare, and hi-hat samples, and a hardware device to select the microphone input. I also use an auto-leveler with a follower, threshold, and sustainer to amplify the vocals and filter out noise. To separate the sounds, I use a sound key filter with a bandpass, low pass, and high pass, and use logic to exclude certain combinations of triggers. Finally, I use a quantizer in trigger mode to put all the triggers on the grid. Overall, this process requires a lot of tweaking and filtering to find the right sweet spot, but it can be done in Bitwig Studio with some experimentation.
You can watch the Video on Youtube - support me on Patreon
In this video, I demonstrate how to convert the input from a microphone into MIDI notes in Bitwig Studio, a popular digital audio workstation. This technique is particularly useful for beatboxing and triggering samples.
I start off by recording the microphone input and converting it into MIDI notes to trigger random samples. I use a trigger function to set off the kick drum, snare, and hi-hat samples, and a separate hardware device to select the microphone input.
To enhance the vocals and filter out any noise, I use an auto-leveler with a follower, threshold, and sustainer. It's a bit of a delicate process to separate the sounds, but I do it by using a sound key filter with a bandpass, low pass, and high pass. I also use logic to exclude certain combinations of triggers.
At the end, I use a quantizer in trigger mode to put all the triggers on the grid. It's important to note that this process requires a lot of tweaking and filtering to find the right sweet spot, but with a bit of patience and experimentation, it's entirely possible to achieve in Bitwig Studio.
I also discuss how this process is similar to creating music in Ableton Live. For instance, I talk about how you can use beatboxing to trigger samples in Ableton Live. It's a challenging process that requires a lot of trials and error, but with some effort, it's possible to achieve similar results in Bitwig Studio.
The method I've described uses a form of AI to separate the different sounds, like the kick and snare, which are created from the mouth during beatboxing. This process goes beyond simple filtering, and takes advantage of the advanced features available in Bitwig Studio.
Questions & Answers #
Maybe you dont watch the video, here are some important takeaways:
What is the primary objective of the video? #
The primary objective of the video is to teach viewers how to convert the input from a microphone into MIDI notes in Bitwig Studio for beatboxing and triggering samples. This includes recording the microphone input, using a trigger for kick drum, snare, and hi-hat samples, and using a hardware device to select the microphone input.
What tools or techniques are used to achieve this objective? #
Several tools and techniques are used in this process. An auto-leveler with a follower, threshold, and sustainer is used to amplify the vocals and filter out noise. A sound key filter with a bandpass, low pass, and high pass is used to separate the sounds, and logic is used to exclude certain combinations of triggers. Lastly, a quantizer in trigger mode is used to put all the triggers on the grid.
What challenges might one encounter when trying to separate different sounds during the beatboxing process? #
Separating different sounds during beatboxing can be challenging because it requires a lot of precise filtering. It's hard to distinguish between different sounds like kick drum, snare, and hi-hat that are made from the mouth during beatboxing. This is why the process involves using AI to separate these sounds, which goes beyond simple filtering.
What does the comparison with Ableton Live reveal about the process? #
The comparison with Ableton Live reveals that similar techniques can be applied in different digital audio workstations. However, it's noted that this process may be more difficult and unstable in Ableton Live, requiring numerous attempts to get a good take. The video demonstrates that the process can be more streamlined in Bitwig Studio with some tweaking and experimentation1.
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.
[00:00.000] Hey folks, welcome back to another video. Today it's about how we can do Ableton Live
[00:05.760] stuff in Bitwig Studio Part 3000. Today it's about, yeah, recording your microphone
[00:13.680] and then convert this into MIDI notes to trigger some random samples. Here's someone
[00:18.880] that is an Ableton Live. So he does some beatboxing and then he triggers the kick, snare,
[00:31.440] and the hi-hat on some trump sampler and he wrote, I got beatboxing to rock and stock Ableton.
[00:38.080] It's super unstable but I got one good take. So he probably recorded this 1500 times
[00:45.360] because it's really hard as you can see soon to separate each of these different sounds. We
[00:52.480] have to do a lot of filtering and I think the native plug in your chord for clear also, I don't know
[01:02.320] what's the right name is. They did this with AI to separate basically kick sounds from snare sounds
[01:08.640] from your mouth so they use some recognition there, not just simple filtering. But you get very
[01:15.840] far with some band pass filtering, high pass filtering. I want to show you this and I see it's
[01:20.400] already downloaded to you on no idea why. But yeah, that's your usual Bitwig Studio question you get
[01:29.920] or I get all the time how you can do the same stuff in Bitwig Studio that people do in Ableton Live.
[01:36.800] That's how it is, right? So here we go. So I show you this here. I give you a rough idea how to
[01:42.320] approach this in Bitwig Studio. So I've already here an XO sampler. It's just a VST device. You can
[01:48.560] also use a drum rack or a drum machine whatever you want to use. I use here just a VST with a kick,
[01:57.040] snare, or a rim shot in the hi-hat sample. And we want to trigger this now with the microphone.
[02:05.440] So I would use in this case here a node grid. You can use whatever you want. You can also use a
[02:12.800] pulley grid if you want to. And we just scrap that because we have no node inputs when we have
[02:19.760] here some node outputs. So I use a trigger and I want to trigger the kick drum which is on C1.
[02:28.720] So we do this. This is the kick drum. That's the snare here, do you want?
[02:39.040] And we want to use here the hi-hat. That's the easy part, right? So now we want to use the
[02:48.720] microphone input. So we use a hardware device, hardware in. You can select your microphone,
[02:54.720] you can see also the level here already. And we probably want to use an oscilloscope.
[03:00.800] So we get here some information about audio input. So the first problem we have to
[03:08.160] tackle is that we have maybe fluctuating level here which is not really nice.
[03:16.480] But you can work with that. You can increase the volume or the amplifier on your audio sound
[03:24.640] code but we can also implement something like an auto-leveler. Like I showed you in some of my
[03:30.720] recent videos so we needed to buy it here and we need a follower. So we follow the audio input
[03:38.160] here, ease the rise and the fall and divide or use this, divide this through the audio signal
[03:48.480] and we get here some kind of amplified signal. The only problem now is that we also
[03:53.920] amplify really quiet signals like the noise floor. You can see it's getting slowly louder.
[04:06.160] Right, so you have noise then that's get amplified and you don't want to have this,
[04:10.480] you want to have some kind of threshold in there. So we implement your small little threshold for
[04:18.400] the noise floor and you need a sustainer all here. So you want to sustain the level here
[04:31.440] when it's got logic here, this one and the constant. So this is basically our threshold here.
[04:45.040] Right, and you pull this down to maybe, I don't know, 0.10. So every time the level goes below
[04:58.240] this white line, this threshold, every time the level goes below that we want to hold the last
[05:06.320] known level. So we do this here with this, right? So every time this follower goes below this level,
[05:16.080] it sends out basically a trigger.
[05:23.040] Something like this, every time we go above this, we have no trigger and when we go below this level,
[05:29.520] you have a trigger, right? So this means every time we go with the vocals above the threshold,
[05:38.000] it uses the values from the vocals to define the volume and every time we don't speak, so the volume
[05:44.480] goes below the threshold, we use the last known level and just hold that with the level sustainer
[05:52.240] here. Okay, so this is basically our small little auto-leveler kind of tool device and we can
[06:02.160] use this now as an input. I can't beatbox, I have to tell you that, right? Or you can give you
[06:13.760] some rough ideas. So you can change the threshold to your kind of level, to your level of your
[06:20.080] microphone and, you know, get some stuff in there. What you also can do is maybe already filter
[06:28.880] the input with the low pass and high pass, get some plops out, something like this, maybe
[06:41.680] everything that's below what it hurts just to get the rumble and you know all that stuff out,
[06:48.240] we don't need and maybe everything above 10k, I have no idea. So we have already some
[06:59.760] pre-filtering here happening. So now we need to separate the sounds, so the kick, kick,
[07:07.280] kick, kick, kick, kick, kick, kick, kick, kick, kick, kick, kick, you know that stuff.
[07:14.560] And we want to use probably sound key. Sound key filter here in the grid is pretty nice because
[07:21.200] it gets pretty steep filters here 24 dB per octave band pass, low pass and high pass. So pretty steep,
[07:30.080] pretty clean. So perfectly nice for that. We want to also remove the pre-code for the pitch,
[07:37.200] so we don't change the frequency when we hit accidentally the keyboard and we want to probably
[07:43.120] use your LP8. So we filter out everything that's below 150. I don't know. You have to treat this
[07:52.400] to your voice. That's the main takeaway here. Still a scope, so we can see what's going on here
[08:00.320] with that. So now output when we want to trigger the hi-hats, that's good. So we get the kick drum
[08:13.360] still there. We can also use the vectors even steeper. We can use two of them,
[08:25.520] something like this. Okay, nice. And then we want to use a follower, probably.
[08:34.400] Get some kind of signal. And then we want to, and then we want to implement some kind of
[08:51.600] threshold, that's a threshold constant. So every time we surpass 0 to 10, we want to trick
[09:02.000] out something. And we do this here. Logic, bigger than this is bigger than that, right? So we
[09:15.680] get the signal out here. So that's basically the kick drum for you.
[09:23.600] Let's put this over here.
[09:42.160] So you can play around you with the full time to get the right amount and the constant.
[09:46.320] You know, they have to tweak the three awards also here with the filters. If you're a kick drum
[09:51.920] sounds from your mouth are way deeper or way higher. Of course, you can treat this here, right?
[09:57.760] So you have to treat this to your voice. So that's the kick drum. Then we go here to the middle part,
[10:05.120] which is the snare drum. So here we probably want to use some bandpass filtering,
[10:12.160] EP8. So we get no signal at all, which is very nice because we are at 150
[10:20.000] hertz here. Maybe go higher or maybe let's go and use a spectrum here.
[10:26.640] We have to make a lot of analyzing here. I can see. So we got 2.5K.
[10:36.160] And here we use the same thing. So we have no signal there. Nice.
[10:51.920] It's made to two steep. Maybe let's go to four.
[10:58.640] So we get something there. So it's a lot of tweaking. That's basically what he said.
[11:05.920] With this line here, I got beatboxing to work and stockable, but it's super unstable.
[11:11.440] It's a lot of work to find the right sweet spot.
[11:26.000] So that's kind of works. And here we do the same for the high pass.
[11:32.800] I pass 8. Everything that's above 2K.
[11:38.640] We get a lot of strong signal here. A lot of strong signals.
[11:45.440] Nice.
[11:51.520] Okay, so it kind of works. It's not perfect, right? So you have to play around with the frequencies,
[11:57.200] as I said. But there's also something you can do to make this a bit better.
[12:02.240] You can bring in some logic stuff. So sometimes when you do a sound here, like
[12:08.720] so you trick out the kick in the snare and the high head at the same time.
[12:15.520] Because you make such a variety of frequencies with your mouth so you can bring in a bit of logic.
[12:21.600] You can say logic and not. So every time you trick out the snare drum,
[12:42.080] or you want to trick out the kick drum, you can only trick out the kick drum when the snare drum
[12:50.480] is not not triggered, right? So you basically exclude this. So when you make
[13:00.880] make this kind of sound, but at the same time this filter picks up some kind of noise below
[13:07.520] the threshold. It doesn't get triggered because you already trigger the snare. So you can
[13:12.880] exclude certain combinations, right? So you can play around with this kind of logic stuff.
[13:20.560] So it kind of works. It's not perfect, but it's something you can work with, right? And you can
[13:33.280] also see you can do a lot of logic stuff with this here. Also you can say then, of course,
[13:38.880] only trigger. Only trigger the high head here when the snare is not triggered.
[13:52.560] Okay, so maybe you introduce your gate length to that. So you make the gate at least a certain amount
[14:04.400] long. Kind of works, works nice actually for the first try. So yeah, you sell key filters,
[14:24.080] they are pretty great, pretty clean. Use logic to exclude certain combinations of triggers.
[14:31.920] And yeah, that's basically my main tip from that. Another idea could be to
[14:44.640] quantize this. So you can use a quantizer. Use this quantizer and trigger mode. Put this in here.
[14:53.600] And then
[15:04.880] use your triggers and go to 16 nodes.
[15:11.200] So now all your triggers are basically on the grid.
[15:13.840] Let me see when I use your instrument track. And I want to record input of the node criteria, right?
[15:26.720] I can't beatbox, guys. So in here, you can see it's... Where is it? Oh, yeah. It's absolutely not on the grid.
[15:49.520] But you get the idea. You get the idea. It kind of works. I put the preset here on the
[15:58.240] scripture below so you can download it and can start from this if you are too lazy to replicate
[16:03.520] this in the grid. Maybe I use... I don't know how to call this. But I put this into the scripture
[16:18.320] below to get you started. And yeah, that's it for this video. Leave a like if you've found this
[16:24.640] helpful. Subscribe to the channel to get more of these kind of Abelden live, how to do Abelden
[16:31.840] live stuff in bitrate videos in the future. Thanks for watching and see you in the next video. Bye.