WEBVTT 00:00.000 --> 00:07.440 And the other reason there are yes, but I'm not in so many here. 00:07.440 --> 00:11.440 I'm one of the organizers of the debut, but I'm also a hobbyist musician. 00:11.440 --> 00:13.680 So I'm really excited to present something today. 00:13.680 --> 00:18.320 And of course, since I am in musician, I pretty much had to bring a beautiful controller 00:18.320 --> 00:20.280 to act as the clicker for my life. 00:20.280 --> 00:24.880 So it your own dog food as they sing. 00:24.880 --> 00:30.240 So just as a second, me, so that I know that I don't run out of time. 00:30.240 --> 00:36.120 Just let me look for double-check when I'm supposed to finish because I want to. 00:36.120 --> 00:36.640 OK. 00:40.120 --> 00:46.640 So yeah, today I've been talking a bit about virtual orchestration using a free and open-source 00:46.640 --> 00:48.920 software and a word of advice. 00:48.920 --> 00:52.280 I mean, a couple of years ago, in open media, I did a presentation where I said, 00:52.280 --> 00:55.680 become a rockstar using free and open-source software. 00:55.680 --> 01:00.280 And that came out nicely, but it was also very clickbaiting because I'm not a rockstar 01:00.280 --> 01:01.280 idol, as you can see. 01:01.280 --> 01:03.920 That's pretty much how many people listen to me monthly. 01:03.920 --> 01:05.040 And one of them is my mother. 01:05.040 --> 01:08.680 So it doesn't really count. 01:08.680 --> 01:13.320 And so you may be asking, but you may be saying that I don't even look like a composer myself. 01:13.320 --> 01:15.960 And you're probably very right. 01:15.960 --> 01:18.120 But I do like to write music. 01:18.120 --> 01:19.280 I do like to play guitar. 01:19.280 --> 01:22.960 I do like to write and record and distribute music myself. 01:22.960 --> 01:27.200 And especially pretty much 90% of what I write as a orchestra lowly 01:27.200 --> 01:30.120 and elements in there because I love progressive rock. 01:30.120 --> 01:30.560 I like it. 01:30.560 --> 01:31.880 We made it like hard rock. 01:31.880 --> 01:35.360 And whenever an orchestra is involved, they will be happy. 01:35.360 --> 01:37.880 And I love soundtracks and classical music. 01:37.880 --> 01:40.560 So orchestra music is very much of my mind. 01:40.560 --> 01:45.080 And so I wanted to come up with the workflow that worked for me in order to write 01:45.080 --> 01:48.360 for orchestral parts as well as part of my work. 01:48.440 --> 01:52.480 So this is just going to, I'm just going to share what's worked for me, which doesn't 01:52.480 --> 01:55.160 mean that it will be the best approach out there. 01:55.160 --> 01:59.840 But just I share basically some guidelines about how I like to do things. 01:59.840 --> 02:01.600 And I love for some alternatives too. 02:01.600 --> 02:07.440 Just so that in case you want to do something similar, you have some tips here and there. 02:07.440 --> 02:13.440 And basically, when we talk about virtual orchestration, first of all, we are talking about 02:13.440 --> 02:18.320 scoring parts, writing parts, and then having some virtual instruments that can turn 02:18.400 --> 02:23.840 them into actual sounds mostly because an orchestra can have a ton of different instruments 02:23.840 --> 02:27.120 instruments that can play, be played by a ton of people. 02:27.120 --> 02:30.880 And of course, I don't know about you, but I don't know 100 people that would be willing 02:30.880 --> 02:36.360 to play my orchestra music, which means that somehow we have to figure out a way to write 02:36.360 --> 02:37.800 these things on a laptop. 02:37.800 --> 02:42.480 And then the laptop will play everything for us. 02:42.560 --> 02:48.640 And of course, the quality of the samples does play a role, of course, the best, if the samples 02:48.640 --> 02:55.920 are more realistic, the more realistic, the more realistic the orchestra will be in general. 02:55.920 --> 03:01.200 But, I mean, in general, a proper arrangement can help, a bit of misdirection can help, 03:01.200 --> 03:04.960 especially when you use orchestras with other things like, for instance, rock or metal 03:04.960 --> 03:08.680 and something like that, you can hide something that are not as realistic. 03:08.680 --> 03:13.240 So, there are many things that you can do in order to make it sound more realistic than 03:13.240 --> 03:19.080 it actually is even using something that is not super expensive as very expensive libraries. 03:19.080 --> 03:23.080 And when we talk about our orchestras, we are talking again about a lot of instruments, 03:23.080 --> 03:26.880 but luckily for us, most of these instruments are also grouped, which means that we don't 03:26.880 --> 03:30.360 need 100 virtual instruments, we will need less than that. 03:30.360 --> 03:35.520 For instance, violins are always grouped, same things for violas and so on and so forth. 03:35.520 --> 03:40.080 Most importantly, also, the different instruments are also grouped by their own in terms 03:40.080 --> 03:45.920 of their own characteristics, where woodwinds blasts, your percussion strings and so on and so 03:45.920 --> 03:46.920 forth. 03:46.920 --> 03:50.040 So, a bit of knowledge there can help. 03:50.040 --> 03:57.600 And as I mentioned, the idea is to go from these to these, because I don't look like 03:57.600 --> 04:02.040 it, but I am getting older and so that's basically what I wanted to do. 04:03.000 --> 04:07.320 I think that the first way to start, I mean, even before we get to the technical part, 04:07.320 --> 04:11.880 I think that one important point when you want to start writing orchestras music is that you 04:11.880 --> 04:16.520 have to do a bit of homework, which does not mean that you have to go to the museum instruments 04:16.520 --> 04:21.560 of hearing process and learn all of the instruments out there like they are your own, 04:21.560 --> 04:28.760 but the museum is fantastic, I suggest, visiting it, but in general, I don't even suggest 04:28.840 --> 04:32.840 reading all of these books because I certainly didn't, I scheme to them, but in general, 04:32.840 --> 04:38.840 have some basic understanding of how the orchestral groups work, how the different instruments work, 04:38.840 --> 04:44.840 the different textures that instruments can produce and which one sounds well with each other, 04:44.840 --> 04:45.800 and so on and so forth. 04:45.800 --> 04:50.360 So, the more, for instance, classical music or sound tracks that you listen to, the more you'll 04:50.360 --> 04:55.080 get accustomed to that, because there are a lot of things that a lot of shortcuts that you can 04:55.080 --> 04:59.000 take in order to sound more realistic. And so, I mean, again, you don't have to watch the 04:59.000 --> 05:03.560 riddle of these books, but there are many videos, for instance, on YouTube that will give you 05:03.560 --> 05:08.920 a crash course about why violins and flutes sound well together, for instance, or the concept of 05:08.920 --> 05:15.080 dubbing and so on and so forth. So, again, do a bit of homework before, not necessarily to study 05:15.080 --> 05:20.360 music in detail, just have a basic understanding of what you're doing, and then when I started 05:20.360 --> 05:27.000 basically, oh, went too far. When I started working on an orchestral set-up, five or six years ago, 05:27.000 --> 05:33.000 I started from this free book by Peter Schafter, who did basically a complete set-up using 05:34.280 --> 05:40.200 new score, Jack, and so on and so forth. It is an excellent book that I suggest, as a reference, 05:40.200 --> 05:44.280 it was a bit too complicated for me at the time, so I basically decided to take a step back and start 05:44.280 --> 05:49.960 from the basics. And so, first of all, how do I write the different parts for the different instruments, 05:50.040 --> 05:53.960 and the first one that I started from was called Lily Pond, that is basically like a 05:53.960 --> 05:59.640 latte kind of programmatic language, where you write a program, you write music as a fitware, 05:59.640 --> 06:03.400 a computer program, and then you compile it, and then you generate beautiful music shift, 06:03.400 --> 06:07.720 for you, it can generate me, for you, and so on. And this is something that I still like a lot 06:07.720 --> 06:12.440 too quickly, write down ideas, but if you want to write something as complex as an orchestral score, 06:12.440 --> 06:17.560 which can take a lot of different things together, using a what you see, what is what you get, 06:17.560 --> 06:22.280 kind of approach works much better, which means that I eventually fell on music or instead, 06:22.280 --> 06:27.560 which is the basically the defecto standard for, especially in the open source word, for this kind of 06:28.760 --> 06:34.760 notations, so often, especially for orchestral music. And I use this one, because even though I'm 06:34.760 --> 06:40.920 not that good at reading music, I prefer writing music that way, I personally do, but if, 06:40.920 --> 06:44.840 for instance, you are much more comfortable playing a keyboard and playing the part of self, 06:44.840 --> 06:49.960 for you prefer piano roles, there are definitely plenty of solutions there. The important part is 06:49.960 --> 06:54.600 that one way or the other, you have to write the different parts for fruits, violins, 06:55.480 --> 07:03.400 trumpets, or whatever, and to get basically the MIDI files that then you will render later on. 07:04.200 --> 07:09.640 And music was the best solution for me there, because it also comes with some templates for 07:09.640 --> 07:15.400 orchestral sections, like a full symphony orchestra and stuff like that. And so once I got the 07:15.400 --> 07:19.960 notation part done, it was a matter of how can I get some better sounds, because with music or 07:19.960 --> 07:26.600 also the box, I can get decent sounds. I mean, it gives me a good preview of what a few, 07:26.600 --> 07:32.040 flute sounds like or violin and so on. If I play everything together, it sounds, it gives you 07:32.040 --> 07:37.720 an idea of what I want to do, but it's not super realistic, we need something better than that. 07:37.800 --> 07:43.160 And there are different alternatives in the open source and free space. I'll mention some of them 07:43.160 --> 07:46.680 in this presentation as well. I also towards the end of the presentation, I'll get back to 07:46.680 --> 07:52.840 music sounds. There are many options there. The one that I stumbled on and that she was for a 07:52.840 --> 07:57.960 long time was virtual playing orchestra, which is a free orchestra sample library written in 07:57.960 --> 08:02.600 Sporzando format. I don't have much time to talk about Sporzando in general, but Sporzando 08:02.600 --> 08:09.320 is basically a text based format that describes how you need to use and play a collection of 08:09.320 --> 08:15.640 samples that you have. So if somebody recorded a ton of samples of a flute playing, using Sporzando, 08:15.640 --> 08:21.640 you can use those samples so that people can play a flute programmatically, basically. 08:22.200 --> 08:27.560 And basically, the author of virtual playing orchestra did the collage of all the best orchestra 08:27.560 --> 08:34.280 libraries out there, all the instruments that he liked the most, he picked from different places. 08:34.280 --> 08:38.440 And then he put them all together and then created difference for some of the files for different 08:38.440 --> 08:42.760 things. So rather than a single big piece for some of the file for your orchestra, you have 08:42.760 --> 08:48.520 one for strings, one for flutes, one for a single flute, one for strings with articulation changes 08:48.520 --> 08:53.960 using key bindings. So you have really a lot of flexibility to handle each individual instrument 08:54.920 --> 09:00.040 pretty much any way that you own. So and these sounds are pretty nice as well, I think, 09:00.040 --> 09:07.720 for at least I like them a lot, not all of them, but I like them. Which then meant that I need 09:07.720 --> 09:14.760 to move one more step forward because I knew how to notate parts. So I knew how to write parts. 09:14.760 --> 09:19.560 I knew how those parts could be rendered, but then I needed to put all of these into the 09:19.560 --> 09:24.600 framework of something that I could work on. So basically like create a project within which 09:24.600 --> 09:29.560 I could start working on my orchestra music altogether in order to come up with the song or a piece 09:29.560 --> 09:34.760 of music at the end. And I use our door a lot for pretty much everything, mixing, mastering and 09:35.720 --> 09:40.520 and everything. And so I wanted something that I could do there. And luckily for me, Michael Wheelies, 09:40.520 --> 09:45.720 which is a very smart developer and an excellent musician, all of basically shared 09:46.520 --> 09:52.120 a template where we basically prepare the template for our door, which we different tracks for 09:52.120 --> 09:57.000 all the different instrument groups using virtual orchestra, a guitar playing orchestra itself, 09:57.000 --> 10:03.000 which was basically exactly what I needed. Using basically for all the instances of 10:03.000 --> 10:08.680 a guitar playing orchestra you use in the fits, which is a plugin that plays for sound or 10:09.560 --> 10:15.000 fives and is very light. And it has a couple of interesting things that I'll mention 10:15.000 --> 10:19.880 more in detail in the next slide because it's easier from a visual perspective. But then the idea was 10:19.880 --> 10:24.200 that I took this template and I tweaked it a bit from my own needs, so cutting the stuff that I 10:24.200 --> 10:30.760 didn't need, like trying to group instruments, reduce the number of instances, tweaking a bit 10:30.760 --> 10:36.760 how I did articulation. So I'll mention it later. And then as I mentioned, I do a lot of 10:36.920 --> 10:42.600 cross genres. So I do for instance a metal and orchestra, which means that in my template I also 10:42.600 --> 10:48.680 need templates for piano, guitar, drums and whatever in case I end up needing them. So I enrich 10:48.680 --> 10:54.520 these templates for my own needs. And one thing that the template gave me out of the box, though, 10:54.520 --> 11:00.920 and was really nice was basically an orchestra layout. And basically, Michael used the stereo space 11:01.000 --> 11:06.840 so panning to place instruments on either side of the orchestra spectrum. And then used the 11:06.840 --> 11:12.680 three different levels of reverb using dragonfly reverb to give them depth, so that some instruments 11:12.680 --> 11:17.640 sounded more in the back, some instruments sounded more in the front. And so the two things together 11:17.640 --> 11:23.560 really gave me a really nice orchestra layout out of the box, which was really helpful for me. 11:23.560 --> 11:28.360 Also because I used the same reverbs for all the instruments that I added, so that for instance drums 11:28.360 --> 11:33.320 would sound as if they were in the same room as the orchestra in an option. So something really 11:33.320 --> 11:39.240 nice. And at this point, basically, I added everything that I needed, I just needed to come out with 11:39.240 --> 11:46.200 a workflow to go from A to Z, basically. So using new score for scouring and then figure out how to 11:46.200 --> 11:51.320 bring everything in order to start working on it. And basically, it is a multi-step approach. So 11:52.520 --> 11:56.840 again, I start scoring the parts in new score, the older cats, all parts are written there. 11:57.320 --> 12:03.080 Then once I'm happy with everything, I export all the individual tracks, which is usually an instrument 12:03.080 --> 12:09.960 group into a MIDI file. I create a new order project using that modified template. And then each 12:09.960 --> 12:15.080 of the MIDI track, each of the MIDI files goes to the right MIDI track. So I have a track for the 12:15.080 --> 12:20.600 flutes. The MIDI that I wrote for flutes in new score is ending there, so that it will be played by 12:20.600 --> 12:28.040 virtual orchestra. We're playing orchestra using the flutes sound, basically. Then there is one additional 12:28.040 --> 12:32.200 step that I mentioned more in detail in a second that is tweaking articulation changes. And 12:32.200 --> 12:37.160 by articulation changes, I mean that an instrument that is not a single thing. For instance, when you 12:37.160 --> 12:42.120 play a violin, you have a violin sound, but my violin could be pitch-sicato, could be tremolo, 12:42.120 --> 12:47.400 could be staccato. And those are typically all different samples because they have different sounds. 12:47.400 --> 12:52.440 And these are typical articulations, which means that some of you need the way to switch from one 12:52.440 --> 12:57.320 to the other, even though you are on the same instrument, basically. So I added something to handle 12:57.320 --> 13:02.920 that. Then if I'm working on something that is not just a orchestra, I add all of the different 13:02.920 --> 13:08.680 parts, I record drums, guitars or whatever. And then once I have everything in the same place, 13:08.680 --> 13:12.840 in order to reckon basically work on the mixing, fine tune everything, because at this point, 13:12.840 --> 13:17.000 I have, I don't know, 40 different tracks and you have to make sure that they will sound 13:17.080 --> 13:22.280 pretty much nicely. You have to EQ everything nicely. And this is usually the one that takes 13:22.280 --> 13:27.800 me the most. I mean, you are pretty much all familiar with that. And I mentioned articulations 13:27.800 --> 13:34.120 as quite important, mostly because the way that my workflow works, musical and vitro-paying 13:34.120 --> 13:39.880 orchestras are handled articulations differently, because if with musical, you've rewrited, 13:39.880 --> 13:46.440 tremolo parts, the musical will rise to the same node in me the 20 times, for instance. That's 13:46.520 --> 13:52.360 not something that I want. I want a single MIDI note that uses a sample that has a tremolo sound 13:52.360 --> 13:58.040 instead, which is something that vitro-paying orchestras provides. So typically, this means that when 13:58.040 --> 14:03.880 I write the parts in mu score, I strip all articulations from there. I basically remember where 14:03.880 --> 14:12.120 they're supposed to be. And then basically I modify the vitro-paying orchestras files for articulation 14:12.200 --> 14:17.400 changes, because by default, vitro-paying orchestras are using uses keyswitches. So basically, 14:17.400 --> 14:23.480 if a violin is in this range, it says press this key here, which is outside of the range 14:23.480 --> 14:27.160 of the violin to switch to tremolo. And this is not something that I wanted, because I didn't 14:27.160 --> 14:33.640 want to pollute the score itself. I wanted something like custom MIDI commands to do the same. And so 14:33.640 --> 14:40.920 I basically chose a random MIDI command there and I mapped it to a few things. So if a 14:40.920 --> 14:48.040 MIDI-14 is in the range 20-39, then it means that I won't tremolo, which does mean that I have to do 14:48.040 --> 14:53.560 some manual tweaking once I put everything in order, because those tweaks are not in mu score, 14:53.560 --> 14:58.360 they are going to be added manually to order, but it works for me. And this is all open, 14:58.360 --> 15:04.360 so of course, if you want to use that, there is a repo over there. And then I don't know if I'll 15:04.360 --> 15:09.560 let time to actually make you listen to some of the words, but if you're curious later on, 15:09.560 --> 15:14.920 for instance, this is one of the first attempts where I tried to do all of these in the same piece. 15:14.920 --> 15:20.840 And this in particular was a classical piece, so it was a symphonic poem, like more like a soundtrack, 15:20.840 --> 15:26.360 so a completely orchestral piece. So it sounded, in some parts it sounds better than others, 15:26.360 --> 15:32.760 but I think it sounded decent enough. In general, here I was a bit more ambitious, because basically 15:32.760 --> 15:38.600 I hate, all I want for Christmas is you by Mariah Carey. And so for Christmas, I decided to turn 15:38.680 --> 15:44.120 it into a Viking metal song, and with the full orchestral thing behind it. So basically, 15:44.120 --> 15:51.560 it's like hearing demoborgive playing Mariah Carey's song, basically. And so this is also available 15:51.560 --> 15:58.120 as a video, if you want, and so in this case, the orchestra is sometimes it's alone, sometimes it's 15:58.120 --> 16:04.040 with a full blasting metal out there, and it sounds nice as a combination of that in general. 16:04.600 --> 16:11.160 As I mentioned in general, I tried to write a lot of orchestral music and rock together, 16:11.160 --> 16:16.600 in for instance, Klopatra's and Nipi is actually a song that lasts about 30 minutes, 16:16.600 --> 16:24.360 a symphonic progress, that emerges a lot of classical and rock elements together, so it was like 16:24.360 --> 16:31.880 a revolution from what we saw before, everything in the same place. And as I mentioned, 16:32.600 --> 16:38.440 virtual playing orchestra is not the only option that you have. One popular option, in general, 16:38.440 --> 16:46.040 is BBC symphonic orchestra discover, which is not open to us because it is made by a company 16:46.040 --> 16:53.800 that is called Spitfire Audio that releases proprietary virtual instruments, but this particular 16:53.800 --> 17:00.280 BST they offer for free, because it's basically a subset of the much more expensive library that they 17:00.280 --> 17:06.360 sell. So it's not super complete, meaning that all the instruments are there, but only the instrument 17:06.360 --> 17:12.760 groups. So you cannot have a solo flute, for instance, it doesn't have legato, the dynamics are not 17:12.760 --> 17:17.480 great, but in general, what is there does sound really good, which is why a lot of people use it 17:17.480 --> 17:22.600 anyway. And luckily for me, I mean, even though it is a wing, those, and I think, Maconly, 17:22.600 --> 17:27.560 virtual instruments, it does work on Linux, if you use something like Linux, BST, which is what 17:27.640 --> 17:33.640 I personally use, or I upgrade, which is another common option, you do need to make a bit of work 17:33.640 --> 17:39.080 because recently they updated the library and it kind of sucks, but if you basically do some 17:39.080 --> 17:45.880 tweaks to use a previous version, it still works great. And what I did use it for was not to replace 17:45.880 --> 17:51.800 what I've shown you before. So I still use virtual playing orchestra a lot, but now in addition to that, 17:51.880 --> 17:58.200 I also use these other library, typically either for blending purposes or for nitpicking, which instrument 17:58.200 --> 18:03.400 I like, because for instance, virtual playing orchestra, I hate the clarinet sound, and so for clarinet, 18:03.400 --> 18:08.440 I just use BBC Symphony Orchestra, for strings that will use both of them together, because it 18:08.440 --> 18:15.320 gives a more lush sound in general. So it's just a matter of having more options available basically. 18:15.320 --> 18:19.960 And this is what I ended up using in my latest album, which is more of a progressive rock album 18:20.040 --> 18:26.680 with orchestral elements. So in this case, I did use this VPO plus BBC Symphony Orchestra 18:26.680 --> 18:31.400 thing together, and I do like how it sounds. I mean, there are parts that may sound better, 18:31.960 --> 18:37.640 we are still talking about more or less synthetic sounds, but I like the output a lot. 18:38.440 --> 18:42.840 Which makes me to my last point, because I mean, I started talking about new score, and some of you 18:42.840 --> 18:49.160 may know that new score has an excellent set of orchestral sounds as well as parts that are 18:49.160 --> 18:54.760 called music sounds basically. And this is something that they add in a recent version of 18:54.760 --> 19:00.440 new score that is called musical four. The problem with that is that, I mean, in general, 19:02.040 --> 19:06.200 if you look for the samples, especially for new two, and so on, it does sound really great. 19:06.200 --> 19:13.080 So it does sound really appealing, and even though I mean, these new sound is basically all 19:13.080 --> 19:18.760 of your orchestral sounds are free as well. So this sounds great. The problem that I had with it at 19:18.760 --> 19:23.880 the time, and this has changed a bit like this, that to use those, you require an external 19:23.880 --> 19:29.400 application that is called MuseHub, which is basically MuseCore's marketplace. So for licensing 19:29.400 --> 19:34.520 reads, and when you install MuseCore, you don't get these new sounds as well out of the box. 19:34.520 --> 19:39.240 You need to do external application to install these sounds separately. And when I tried 19:39.240 --> 19:43.880 it a couple of years ago, these external application was basically a closed surduplication 19:43.880 --> 19:49.560 that required root access, which basically was a big note from me, and for different reasons, 19:49.560 --> 19:53.960 privacy security and so on. But luckily, they listen to the community, and this is not through 19:53.960 --> 19:59.880 anymore. So this means that this is not an issue anymore. And others, more or less, should 19:59.960 --> 20:03.960 that I have, is that of course, this is only available in MuseCore 4, which is the latest one, 20:03.960 --> 20:09.640 available, and I'm still on MuseCore 3. A bit because I'm lazy, a bit because MuseCore 4 doesn't 20:09.640 --> 20:14.840 support Jack, and so it doesn't work well with the pipe, wire version of Jack that I have. 20:14.840 --> 20:20.120 But I mean, it is definitely on my list to start playing with this soon, because I mean, 20:20.120 --> 20:24.600 this is something that the sounds, it sounds really great. So I want to be able to play with 20:24.760 --> 20:30.120 this a bit as well. And so I want you to know that this exists as well. And I mean, I rushed 20:30.120 --> 20:34.840 a bit through this, because there was a lot to say, and there is also a more than in depth blog 20:34.840 --> 20:39.800 post, if you want to learn to read about it a bit more. It is a bit outdated, because I did 20:39.800 --> 20:44.920 write that blog post last year, and some things changed. Like, will the BBC's discover stuff, 20:44.920 --> 20:51.080 I know the MuseCore sound stuff changed a lot, but there is some information there. And I think 20:51.160 --> 20:55.960 this covers everything, so I think I'm pretty much on time to handle some questions as well. 20:55.960 --> 20:57.960 Okay. Thank you. 21:06.680 --> 21:11.960 The microphone, so everyone can hear. So the first question seems to be at the front, that makes it easy for me. 21:12.680 --> 21:16.840 I like easy, just thank you very much. So that's brilliant. 21:17.560 --> 21:22.520 I'm an open source developer myself, but I'm also working my way to become a professional 21:22.520 --> 21:28.520 organization and scoring composer. So that's exactly, almost exactly what I would have done 21:28.520 --> 21:35.160 if it was not for the requirement of using like really high 11 libraries, which is not easy, 21:35.160 --> 21:40.520 if not impossible to integrate with the inox and rest of the system. Have you tried any of the 21:40.520 --> 21:45.880 other libraries like East West or the other speakers or anything compared to what you can actually 21:45.960 --> 21:49.800 use already in terms of sample quality and so on? 21:52.200 --> 21:58.040 No, I think they got the question because it's also in the audio. So I didn't try any of the 21:58.040 --> 22:03.560 more professional libraries, first of all, because 99% of the times they are not available on 22:03.560 --> 22:10.840 Linux, so they are only available in contact, which is a pain in the parts to set up on Linux and so on. 22:11.720 --> 22:16.360 Even when they are kind of available like the BBC Discovery thing, 22:17.320 --> 22:23.160 using a Windows VST on Linux is still a bit tricky. For instance, what I did scheme to and not go 22:23.160 --> 22:28.440 very much into detail is that for instance, to use that discover library, I'm doing the 22:28.440 --> 22:34.040 all the rendering outside of basically, are do it separately. Basically, I have a card line 22:34.040 --> 22:38.200 since where I load the single instance of BBC Discover because it's heavyweight. 22:38.200 --> 22:44.280 When I want to record floats, I have arduer send me the MIDI float to this card line since 22:44.840 --> 22:48.920 that will be rendered there, it will go back to arduer and that will record the audio there. 22:48.920 --> 22:53.480 So it's a bit of a manual process and it would be even more a manual process, the moment I 22:53.480 --> 22:57.960 switch to more professional stuff and I think that new sounds is probably going to be 22:57.960 --> 23:02.440 the closest that you can get to very professional sounds in a setup like that, but I haven't 23:02.440 --> 23:03.800 tested it myself yet. 23:04.680 --> 23:10.360 Still got time for another question or two? Anyone with speaking after me the other way, 23:10.360 --> 23:17.080 because yes, who is next? I know, totally organized and I'm with brilliant. 23:18.360 --> 23:20.440 Ah, Francesco is coming up. 23:20.440 --> 23:22.600 Damn, no, no, no, no. 23:24.600 --> 23:30.920 Let's see, let's see if I can share a bit of the 23:33.800 --> 23:36.920 C feeds, if the audio is coming true, I'm not sure. 23:42.920 --> 23:44.920 I don't think it is just let me give you a second. 23:50.120 --> 23:54.680 Yeah, I'd rather not need to do the HDMI, so let me see if I if it works that way. 23:55.400 --> 23:58.200 26 years of tech conference, we still can't get the tech working. 24:03.880 --> 24:11.880 Yeah, I don't think this is the sounds, these are the sounds that we're expecting. 24:11.880 --> 24:14.520 So I mean, I don't know, okay, is there is there? 24:18.760 --> 24:21.800 So a bit of an ominous classical part. 24:27.240 --> 24:28.600 Let's get to some more noise. 24:34.360 --> 24:36.360 Yeah. 24:47.720 --> 24:50.440 Again, I'd be too noisy about you or you're under there. 24:50.440 --> 24:52.040 There was our case when I walked through there. 24:52.040 --> 24:53.720 Well done. 24:56.360 --> 24:58.680 I've done this no such thing as too noisy. 24:58.920 --> 25:03.400 Actually, we're like the ones I thank you very much for that. 25:07.080 --> 25:08.280 We all continue. Yes.