WEBVTT 00:00:00.000 --> 00:00:00.964 align:middle line:90% 00:00:00.964 --> 00:00:02.892 align:middle line:90% [SQUEAKING] 00:00:02.892 --> 00:00:04.338 align:middle line:90% [RUSTLING] 00:00:04.338 --> 00:00:05.784 align:middle line:90% [CLICKING] 00:00:05.784 --> 00:00:10.122 align:middle line:90% 00:00:10.122 --> 00:00:12.650 align:middle line:84% MICHAEL CUTHBERT: You've seen that there's 00:00:12.650 --> 00:00:14.642 align:middle line:84% more technical problems than a standard class, 00:00:14.642 --> 00:00:16.309 align:middle line:84% and that means you're in the right place 00:00:16.309 --> 00:00:19.200 align:middle line:84% if you're here for a computational class, right? 00:00:19.200 --> 00:00:24.762 align:middle line:84% This is 21M.383 Computational Music Theory and Analysis. 00:00:24.762 --> 00:00:26.970 align:middle line:84% Or I can't remember if I call it analysis and theory. 00:00:26.970 --> 00:00:28.670 align:middle line:90% Don't hold it against me. 00:00:28.670 --> 00:00:30.090 align:middle line:90% I'm Michael Cuthbert. 00:00:30.090 --> 00:00:31.770 align:middle line:90% I'm the professor of the class. 00:00:31.770 --> 00:00:35.460 align:middle line:84% I'm a professor of music in music and theater arts, 00:00:35.460 --> 00:00:39.420 align:middle line:84% and I do some work with computational analysis. 00:00:39.420 --> 00:00:44.820 align:middle line:84% So what I want to start with-- oh, and that's who I am. 00:00:44.820 --> 00:00:46.880 align:middle line:84% I'm going to get to know who you guys are 00:00:46.880 --> 00:00:49.730 align:middle line:90% on Friday and next Wednesday. 00:00:49.730 --> 00:00:51.270 align:middle line:90% It's very important to me. 00:00:51.270 --> 00:00:55.170 align:middle line:84% I want to really quickly be able to learn everybody by name. 00:00:55.170 --> 00:00:58.490 align:middle line:84% But on the first class, I think you're all here 00:00:58.490 --> 00:01:01.930 align:middle line:84% to figure out if this is the place you want to be before we 00:01:01.930 --> 00:01:03.500 align:middle line:90% get to know everybody else. 00:01:03.500 --> 00:01:06.310 align:middle line:84% So that's why I'm not going to be going around and asking 00:01:06.310 --> 00:01:07.900 align:middle line:84% everybody today, but I'm going to-- 00:01:07.900 --> 00:01:10.030 align:middle line:90% I really want to know that soon. 00:01:10.030 --> 00:01:13.330 align:middle line:84% So the big question for this class the whole semester 00:01:13.330 --> 00:01:18.970 align:middle line:84% is going to be, how do computers hear music? 00:01:18.970 --> 00:01:20.560 align:middle line:90% And what does that even mean? 00:01:20.560 --> 00:01:24.310 align:middle line:84% I'm going to give you an answer, if you care about answers 00:01:24.310 --> 00:01:29.350 align:middle line:84% like this, that you can read, absorb, pack up, and go home, 00:01:29.350 --> 00:01:32.450 align:middle line:84% and you've no need for this class if this is satisfactory. 00:01:32.450 --> 00:01:35.620 align:middle line:84% Computers hear music by using a microphone connected 00:01:35.620 --> 00:01:37.930 align:middle line:84% to an analog to digital converter, connected 00:01:37.930 --> 00:01:40.670 align:middle line:84% to a microprocessor with a specialized codec-- 00:01:40.670 --> 00:01:42.730 align:middle line:84% I can never remember what that really means-- 00:01:42.730 --> 00:01:47.110 align:middle line:84% to quantize the sound into a WAV file, or AAC or something else, 00:01:47.110 --> 00:01:49.537 align:middle line:84% file that we store on a magnetic or solid state disk, 00:01:49.537 --> 00:01:52.120 align:middle line:84% or we store it on the cloud and we store it on somebody else's 00:01:52.120 --> 00:01:55.614 align:middle line:90% magnetic solid state disks. 00:01:55.614 --> 00:02:00.190 align:middle line:90% Not a very, very useful answer. 00:02:00.190 --> 00:02:03.960 align:middle line:84% So what I think we more want to know 00:02:03.960 --> 00:02:06.160 align:middle line:84% is, the computers are hearing music, 00:02:06.160 --> 00:02:08.910 align:middle line:90% but are they actually listening? 00:02:08.910 --> 00:02:14.830 align:middle line:84% And these two terms, hear versus listen, in different contexts, 00:02:14.830 --> 00:02:17.760 align:middle line:84% one of them has higher priority than the other 00:02:17.760 --> 00:02:23.700 align:middle line:84% and implies a more deep, deep understanding, like mom or dad 00:02:23.700 --> 00:02:28.295 align:middle line:84% ever tell you, I know you heard me, but did you listen to me. 00:02:28.295 --> 00:02:31.630 align:middle line:84% We have that kind of distinction built in. 00:02:31.630 --> 00:02:35.700 align:middle line:84% So what we want to know is, when computers 00:02:35.700 --> 00:02:41.040 align:middle line:84% hear music, what kinds of understanding might they have, 00:02:41.040 --> 00:02:45.120 align:middle line:90% might be encoded in that, If. 00:02:45.120 --> 00:02:47.010 align:middle line:84% You feel uncomfortable with the notion 00:02:47.010 --> 00:02:51.550 align:middle line:84% of computers having their own autonomy or something like that? 00:02:51.550 --> 00:02:57.290 align:middle line:84% And the only reason I really care about this question 00:02:57.290 --> 00:03:01.410 align:middle line:84% is that, by understanding how a computer hears, 00:03:01.410 --> 00:03:04.220 align:middle line:84% it might change how we hear, or it 00:03:04.220 --> 00:03:06.740 align:middle line:84% might open up new pieces for us to listen 00:03:06.740 --> 00:03:09.660 align:middle line:90% to that we didn't already know. 00:03:09.660 --> 00:03:13.340 align:middle line:84% And most importantly, it might answer questions 00:03:13.340 --> 00:03:16.430 align:middle line:84% about what it's like for us to be humans. 00:03:16.430 --> 00:03:17.760 align:middle line:90% Any androids here? 00:03:17.760 --> 00:03:19.040 align:middle line:90% OK. 00:03:19.040 --> 00:03:21.560 align:middle line:84% What it's like for us to be humans 00:03:21.560 --> 00:03:24.420 align:middle line:90% who are listening to music. 00:03:24.420 --> 00:03:29.870 align:middle line:84% So here are any six big unknowns for me 00:03:29.870 --> 00:03:31.650 align:middle line:84% that I want to know the answer to. 00:03:31.650 --> 00:03:34.670 align:middle line:84% I'll just say, you turn in on Friday 00:03:34.670 --> 00:03:36.888 align:middle line:84% the correct answer to any of these six questions, 00:03:36.888 --> 00:03:39.180 align:middle line:84% you get an A plus and you don't have to take the class. 00:03:39.180 --> 00:03:41.130 align:middle line:90% But I don't know them. 00:03:41.130 --> 00:03:44.570 align:middle line:90% So why do we listen to music? 00:03:44.570 --> 00:03:46.220 align:middle line:90% Why? 00:03:46.220 --> 00:03:48.600 align:middle line:90% We don't go around all day. 00:03:48.600 --> 00:03:53.210 align:middle line:84% I have my different scents that I smell all day or something. 00:03:53.210 --> 00:03:56.210 align:middle line:84% There are all kinds of things that are enjoyable 00:03:56.210 --> 00:03:59.140 align:middle line:84% but that seem to be different, that we don't do all day long 00:03:59.140 --> 00:04:00.280 align:middle line:90% nowadays. 00:04:00.280 --> 00:04:03.850 align:middle line:84% Why do people choose what musical styles they like? 00:04:03.850 --> 00:04:06.940 align:middle line:84% How does your musical tastes, or our whole culture's 00:04:06.940 --> 00:04:09.080 align:middle line:90% musical tastes change over time? 00:04:09.080 --> 00:04:11.020 align:middle line:84% And some of us will be from different cultures 00:04:11.020 --> 00:04:13.173 align:middle line:84% and have different musical tastes. 00:04:13.173 --> 00:04:15.340 align:middle line:84% What's the difference between a great piece of music 00:04:15.340 --> 00:04:17.000 align:middle line:90% and a mediocre or bad one? 00:04:17.000 --> 00:04:20.740 align:middle line:84% Or do those terms even really exist? 00:04:20.740 --> 00:04:26.500 align:middle line:84% What makes a piece of music beautiful or ugly? 00:04:26.500 --> 00:04:28.220 align:middle line:84% Let's not forget about the physical. 00:04:28.220 --> 00:04:30.130 align:middle line:84% Why do I want to dance or move when 00:04:30.130 --> 00:04:31.870 align:middle line:84% I hear some music, and other ones, 00:04:31.870 --> 00:04:34.810 align:middle line:90% I just want to sit and cry? 00:04:34.810 --> 00:04:38.910 align:middle line:84% So these are the questions that if you answer them, 00:04:38.910 --> 00:04:40.160 align:middle line:90% that would just be astounding. 00:04:40.160 --> 00:04:45.010 align:middle line:84% I think, however, that a one semester, twice a week 00:04:45.010 --> 00:04:48.280 align:middle line:84% class, we're not going to get the definitive answer to them. 00:04:48.280 --> 00:04:51.640 align:middle line:84% So I would like to break these questions into slightly smaller 00:04:51.640 --> 00:04:55.000 align:middle line:84% unknowns that help to answer these questions, 00:04:55.000 --> 00:04:59.280 align:middle line:84% and that we might be able to start making progress on. 00:04:59.280 --> 00:05:01.740 align:middle line:84% What particular musical characteristics 00:05:01.740 --> 00:05:05.160 align:middle line:84% distinguish a folk song from one part of the world from one 00:05:05.160 --> 00:05:06.270 align:middle line:90% in another? 00:05:06.270 --> 00:05:09.850 align:middle line:84% How does the look of a band, its costume, age, 00:05:09.850 --> 00:05:14.380 align:middle line:84% its racial composition, affect the makeup of the audience? 00:05:14.380 --> 00:05:17.700 align:middle line:84% You know I'm not about to be a heavy metal star the way 00:05:17.700 --> 00:05:19.090 align:middle line:90% I'm currently dressed, right? 00:05:19.090 --> 00:05:20.290 align:middle line:90% Well, what does that mean? 00:05:20.290 --> 00:05:22.350 align:middle line:90% Maybe I'm OK at the piano. 00:05:22.350 --> 00:05:24.850 align:middle line:84% I'm actually better at metal than the piano, as you'll hear. 00:05:24.850 --> 00:05:26.010 align:middle line:90% But that's OK. 00:05:26.010 --> 00:05:28.380 align:middle line:90% I'm pretty bad at both. 00:05:28.380 --> 00:05:32.830 align:middle line:84% What sorts of chords are used for relaxation, for rest, 00:05:32.830 --> 00:05:34.360 align:middle line:90% for stop in music today? 00:05:34.360 --> 00:05:36.360 align:middle line:84% And how is this different in different centuries 00:05:36.360 --> 00:05:38.640 align:middle line:90% or different parts of the world? 00:05:38.640 --> 00:05:40.950 align:middle line:84% And what do the lyrics of top selling songs 00:05:40.950 --> 00:05:43.590 align:middle line:84% have in common with those of worst selling songs, 00:05:43.590 --> 00:05:45.420 align:middle line:90% or don't have in common? 00:05:45.420 --> 00:05:49.405 align:middle line:84% And that's a question that I know 00:05:49.405 --> 00:05:51.030 align:middle line:84% that there are people at the big record 00:05:51.030 --> 00:05:55.580 align:middle line:84% labels trying to answer that, because there's money involved. 00:05:55.580 --> 00:05:58.130 align:middle line:84% What rhythms or instruments accompany 00:05:58.130 --> 00:06:01.760 align:middle line:84% texts about sadness or death that appear less often 00:06:01.760 --> 00:06:02.900 align:middle line:90% in moments of happiness? 00:06:02.900 --> 00:06:04.910 align:middle line:84% That's kind of related to the why some music 00:06:04.910 --> 00:06:07.100 align:middle line:90% make me cry and so on. 00:06:07.100 --> 00:06:11.690 align:middle line:84% So five good questions that might help us 00:06:11.690 --> 00:06:14.210 align:middle line:90% answer a big unknown. 00:06:14.210 --> 00:06:17.450 align:middle line:84% What I'm going to do is I'm going to leave three up and take 00:06:17.450 --> 00:06:21.110 align:middle line:84% two off, and say that these three 00:06:21.110 --> 00:06:23.840 align:middle line:84% questions in particular, musical characteristics 00:06:23.840 --> 00:06:27.860 align:middle line:84% distinguishing x from y, chords used at certain points 00:06:27.860 --> 00:06:31.670 align:middle line:84% and change over time, rhythms about, and connections 00:06:31.670 --> 00:06:36.020 align:middle line:84% to emotion, these all might be questions that having 00:06:36.020 --> 00:06:40.500 align:middle line:84% a representation of a musical score, whether it's more modern, 00:06:40.500 --> 00:06:42.240 align:middle line:90% whether handwritten, ancient. 00:06:42.240 --> 00:06:46.160 align:middle line:84% These are questions that might help answer. 00:06:46.160 --> 00:06:49.290 align:middle line:90% So let's just kind of recap. 00:06:49.290 --> 00:06:54.710 align:middle line:84% What are some of the things the computer is good at with music? 00:06:54.710 --> 00:06:56.090 align:middle line:90% It's kind of old slides. 00:06:56.090 --> 00:06:57.290 align:middle line:90% You'll see it's pretty out. 00:06:57.290 --> 00:07:01.700 align:middle line:84% But that we make a lot of music with a computer. 00:07:01.700 --> 00:07:02.540 align:middle line:90% We mix it. 00:07:02.540 --> 00:07:08.710 align:middle line:90% We do re-editing with it. 00:07:08.710 --> 00:07:11.000 align:middle line:84% Computers have absolutely changed 00:07:11.000 --> 00:07:15.220 align:middle line:84% how we distribute music, how we get it, how we access it. 00:07:15.220 --> 00:07:18.700 align:middle line:84% And then very cool apps and algorithms 00:07:18.700 --> 00:07:20.680 align:middle line:84% developed at places like this can 00:07:20.680 --> 00:07:26.320 align:middle line:84% tell what song particularly is some drunken jerk 00:07:26.320 --> 00:07:27.450 align:middle line:90% singing in a bar. 00:07:27.450 --> 00:07:31.378 align:middle line:90% 00:07:31.378 --> 00:07:38.035 align:middle line:84% But until about the last decade and a half, 00:07:38.035 --> 00:07:41.950 align:middle line:84% little had changed since the '70s and '80s in how computers 00:07:41.950 --> 00:07:44.890 align:middle line:90% could be used to analyze scores. 00:07:44.890 --> 00:07:46.250 align:middle line:90% And so things are changing now. 00:07:46.250 --> 00:07:50.550 align:middle line:84% But let me say, where things were, 00:07:50.550 --> 00:07:53.620 align:middle line:84% what people already had for, let's say, text, 00:07:53.620 --> 00:07:56.280 align:middle line:90% that's different from music. 00:07:56.280 --> 00:08:00.510 align:middle line:84% So if you load a text-- we're all computer geeks, at least 00:08:00.510 --> 00:08:01.590 align:middle line:90% a little bit. 00:08:01.590 --> 00:08:03.910 align:middle line:84% I consider that as a term of endearment for myself, 00:08:03.910 --> 00:08:05.020 align:middle line:90% but I hope that's OK. 00:08:05.020 --> 00:08:06.606 align:middle line:90% We're all computer people. 00:08:06.606 --> 00:08:10.335 align:middle line:90% And so we can look at-- 00:08:10.335 --> 00:08:11.950 align:middle line:84% I can use the term string, right? 00:08:11.950 --> 00:08:13.840 align:middle line:84% We can look at a string of text like this, 00:08:13.840 --> 00:08:17.790 align:middle line:84% and you can ask various algorithms that already exist 00:08:17.790 --> 00:08:19.170 align:middle line:84% in any programming language to be 00:08:19.170 --> 00:08:21.450 align:middle line:84% able to identify the spaces, split, 00:08:21.450 --> 00:08:23.620 align:middle line:90% or what's a foreign diacritic. 00:08:23.620 --> 00:08:25.800 align:middle line:84% Cast it to ASCII and see what crashes. 00:08:25.800 --> 00:08:27.160 align:middle line:90% What's punctuation? 00:08:27.160 --> 00:08:29.920 align:middle line:84% And then if you have some kind of font rendering, 00:08:29.920 --> 00:08:32.320 align:middle line:84% you'll know what's going below the line, and so on, 00:08:32.320 --> 00:08:34.350 align:middle line:84% what's a capital letter, all these things. 00:08:34.350 --> 00:08:39.240 align:middle line:84% And this level of low level information about text 00:08:39.240 --> 00:08:42.870 align:middle line:84% really helped to build the next level of information 00:08:42.870 --> 00:08:45.660 align:middle line:84% about text sentence diagramming, understanding 00:08:45.660 --> 00:08:49.580 align:middle line:84% from letters to a sentence level understanding. 00:08:49.580 --> 00:08:53.100 align:middle line:84% And this kind of built up knowledge 00:08:53.100 --> 00:08:56.960 align:middle line:84% has helped to make to the next level, which we might be arguing 00:08:56.960 --> 00:09:02.600 align:middle line:84% today about whether or not there is an actual meaning involved. 00:09:02.600 --> 00:09:03.120 align:middle line:90% I'm hungry. 00:09:03.120 --> 00:09:05.480 align:middle line:90% I want spam. 00:09:05.480 --> 00:09:10.080 align:middle line:84% So can we build these technologies for musical scores? 00:09:10.080 --> 00:09:12.672 align:middle line:90% Maybe. 00:09:12.672 --> 00:09:14.550 align:middle line:90% Maybe, but not yet today. 00:09:14.550 --> 00:09:20.580 align:middle line:84% So this is a class that I hope people 5, 6, 7, 00:09:20.580 --> 00:09:22.380 align:middle line:84% 10 years from now-- hopefully less. 00:09:22.380 --> 00:09:24.370 align:middle line:90% Hopefully, not 30, 50-- 00:09:24.370 --> 00:09:28.310 align:middle line:84% that what you do in this class will help get 00:09:28.310 --> 00:09:32.000 align:middle line:90% to that level of understanding. 00:09:32.000 --> 00:09:34.650 align:middle line:90% I think put it somewhere. 00:09:34.650 --> 00:09:37.880 align:middle line:84% This is a unique, really unique class. 00:09:37.880 --> 00:09:40.460 align:middle line:84% I think that there's about three other classes that 00:09:40.460 --> 00:09:44.990 align:middle line:84% are something like computational music theory offered 00:09:44.990 --> 00:09:46.210 align:middle line:90% in the world today. 00:09:46.210 --> 00:09:50.400 align:middle line:84% The other ones are all doctoral seminars or graduate seminars. 00:09:50.400 --> 00:09:55.410 align:middle line:84% This is a class that, because of MIT's unique ability 00:09:55.410 --> 00:09:58.920 align:middle line:84% to merge the arts and programming, a deep knowledge 00:09:58.920 --> 00:10:02.580 align:middle line:84% of music, and a great culture of collaboration 00:10:02.580 --> 00:10:06.945 align:middle line:84% in solving problems, we can do this at the undergraduate level. 00:10:06.945 --> 00:10:08.320 align:middle line:84% And if you're a graduate student, 00:10:08.320 --> 00:10:11.170 align:middle line:84% we can do even more with you all on that. 00:10:11.170 --> 00:10:15.390 align:middle line:84% So what are some of the technologies we need to build? 00:10:15.390 --> 00:10:17.880 align:middle line:84% Well, here's some simple things that a computer 00:10:17.880 --> 00:10:20.080 align:middle line:84% has to know to do anything with scores. 00:10:20.080 --> 00:10:23.950 align:middle line:84% Start off with D, E, F, G. What's the next note? 00:10:23.950 --> 00:10:24.450 align:middle line:90% AUDIENCE: A. 00:10:24.450 --> 00:10:27.400 align:middle line:84% MICHAEL CUTHBERT: A. Not H, right? 00:10:27.400 --> 00:10:30.640 align:middle line:84% Not even the Germans who have H would go on from that. 00:10:30.640 --> 00:10:33.390 align:middle line:84% So you need to know about these cyclic little things. 00:10:33.390 --> 00:10:34.970 align:middle line:90% Take a look at this next slide. 00:10:34.970 --> 00:10:37.830 align:middle line:90% 00:10:37.830 --> 00:10:40.980 align:middle line:84% Yeah, we got some squirmers in the room. 00:10:40.980 --> 00:10:45.650 align:middle line:90% That 3/4, this does not compute. 00:10:45.650 --> 00:10:48.690 align:middle line:84% So we need to have an idea of meter and things. 00:10:48.690 --> 00:10:52.790 align:middle line:84% And then from there, building up which of these two scores 00:10:52.790 --> 00:10:56.259 align:middle line:84% is more consonant and which is more dissonant. 00:10:56.259 --> 00:10:59.682 align:middle line:90% [MUSIC PLAYING] 00:10:59.682 --> 00:11:03.110 align:middle line:90% 00:11:03.110 --> 00:11:06.530 align:middle line:84% So we need a concept of consonance, dissonance, 00:11:06.530 --> 00:11:08.820 align:middle line:84% and how to bring this to the score. 00:11:08.820 --> 00:11:11.480 align:middle line:84% So we're going to need to step back 00:11:11.480 --> 00:11:14.000 align:middle line:84% from some of even the middle sized questions 00:11:14.000 --> 00:11:15.900 align:middle line:90% and go to even smaller things. 00:11:15.900 --> 00:11:20.810 align:middle line:84% So for instance, let's say this is the Tchaikovsky's 1812 00:11:20.810 --> 00:11:21.540 align:middle line:90% overture. 00:11:21.540 --> 00:11:25.490 align:middle line:84% And let's see if we wanted to understand 00:11:25.490 --> 00:11:30.240 align:middle line:84% how the key relations work, how we move from one key to another, 00:11:30.240 --> 00:11:32.300 align:middle line:84% and how this builds up the form, and the tension, 00:11:32.300 --> 00:11:33.480 align:middle line:90% and the power of the piece. 00:11:33.480 --> 00:11:34.897 align:middle line:84% Well, in order to do this, we need 00:11:34.897 --> 00:11:36.630 align:middle line:84% to know where all the key changes are. 00:11:36.630 --> 00:11:39.290 align:middle line:84% So we have to be able to cut the score into smaller pieces 00:11:39.290 --> 00:11:40.530 align:middle line:90% and identify that. 00:11:40.530 --> 00:11:43.490 align:middle line:84% Now, how might you identify a key change? 00:11:43.490 --> 00:11:45.400 align:middle line:84% Well, you might say, well it starts off 00:11:45.400 --> 00:11:48.130 align:middle line:84% with a cadence of on key, and then it 00:11:48.130 --> 00:11:49.700 align:middle line:84% ends in a cadence in another key. 00:11:49.700 --> 00:11:51.797 align:middle line:84% So now we need to be able to identify cadences. 00:11:51.797 --> 00:11:53.380 align:middle line:84% So we need something that can do that. 00:11:53.380 --> 00:11:56.290 align:middle line:90% Well, this chord moves in. 00:11:56.290 --> 00:11:59.750 align:middle line:84% Five moves to one, so that's a particular type of cadence. 00:11:59.750 --> 00:12:00.760 align:middle line:90% There's that term. 00:12:00.760 --> 00:12:03.280 align:middle line:84% When I say five moves to one, are people 00:12:03.280 --> 00:12:06.700 align:middle line:84% hearing Roman numerals in your head and visualizing it? 00:12:06.700 --> 00:12:09.370 align:middle line:84% So these are the levels of musical knowledge 00:12:09.370 --> 00:12:11.180 align:middle line:84% that we're going to be using a lot here. 00:12:11.180 --> 00:12:14.290 align:middle line:84% And so in order to figure out if a chord is a five chord, or one 00:12:14.290 --> 00:12:17.300 align:middle line:84% chord, or what's in the chord at all, if it's major or minor, 00:12:17.300 --> 00:12:21.070 align:middle line:84% we need a representation of chords, and notes and pitches. 00:12:21.070 --> 00:12:22.900 align:middle line:84% So some of what we're going to be doing 00:12:22.900 --> 00:12:27.320 align:middle line:84% at the beginning of the class is recreating some of these tools, 00:12:27.320 --> 00:12:29.380 align:middle line:84% and then we'll see what other people have done 00:12:29.380 --> 00:12:31.750 align:middle line:90% and go further from that. 00:12:31.750 --> 00:12:36.880 align:middle line:84% So the notion that we don't have all these things 00:12:36.880 --> 00:12:40.700 align:middle line:84% is kind of surprising, because there's a lot of money, 00:12:40.700 --> 00:12:42.300 align:middle line:84% a lot of work that's been out there, 00:12:42.300 --> 00:12:46.800 align:middle line:84% taking some of representation of musical score, whether notation, 00:12:46.800 --> 00:12:49.510 align:middle line:84% or MIDI, or something like that, some of a score, 00:12:49.510 --> 00:12:52.812 align:middle line:84% and playing it really nicely on speakers and things 00:12:52.812 --> 00:12:54.520 align:middle line:84% like that, Garage Band, all those things. 00:12:54.520 --> 00:12:58.150 align:middle line:84% Or if you've used Noteflight, Finale, Sibelius, 00:12:58.150 --> 00:13:00.680 align:middle line:84% Dorico, something like that, Flat IO, 00:13:00.680 --> 00:13:03.600 align:middle line:84% you know that we can take a representation of notes 00:13:03.600 --> 00:13:06.160 align:middle line:84% and turn them into some of musical score. 00:13:06.160 --> 00:13:08.850 align:middle line:84% So I think of that as kind of output. 00:13:08.850 --> 00:13:10.030 align:middle line:90% It's like a printer. 00:13:10.030 --> 00:13:15.010 align:middle line:84% And then a lot of work, especially in research labs, 00:13:15.010 --> 00:13:18.090 align:middle line:84% on taking sound in and converting it 00:13:18.090 --> 00:13:20.080 align:middle line:84% into that kind of digital representation, 00:13:20.080 --> 00:13:21.670 align:middle line:90% that's useful for search things. 00:13:21.670 --> 00:13:25.230 align:middle line:84% But once you have a digital representation of a score 00:13:25.230 --> 00:13:27.520 align:middle line:84% already in your computer, what do you do with it? 00:13:27.520 --> 00:13:28.450 align:middle line:90% How do you analyze it? 00:13:28.450 --> 00:13:29.320 align:middle line:90% How do you work with it? 00:13:29.320 --> 00:13:31.195 align:middle line:84% And there's been a lot less research on this. 00:13:31.195 --> 00:13:34.360 align:middle line:84% So it's like we've been spending all our time building printers 00:13:34.360 --> 00:13:34.930 align:middle line:90% and-- 00:13:34.930 --> 00:13:36.930 align:middle line:84% I don't know how do we get things in-- scanners, 00:13:36.930 --> 00:13:41.610 align:middle line:84% and much, much less time building our actual CPUs. 00:13:41.610 --> 00:13:43.950 align:middle line:84% Nobody has a CPU that looks like that anymore. 00:13:43.950 --> 00:13:47.300 align:middle line:84% Time to update your slides, prof. 00:13:47.300 --> 00:13:49.910 align:middle line:84% So one of the things that we're going 00:13:49.910 --> 00:13:52.100 align:middle line:84% to be using a lot in this class is the toolkit 00:13:52.100 --> 00:13:54.770 align:middle line:84% that I and many other people at MIT 00:13:54.770 --> 00:13:59.520 align:middle line:84% have used to do these kind of manipulations. 00:13:59.520 --> 00:14:03.650 align:middle line:84% And I think it's correct to say it's 00:14:03.650 --> 00:14:07.310 align:middle line:84% the most used toolkit in the world for working 00:14:07.310 --> 00:14:09.260 align:middle line:90% with musical scores. 00:14:09.260 --> 00:14:12.360 align:middle line:84% It's Python, free and open source. 00:14:12.360 --> 00:14:15.500 align:middle line:84% And sometime over the next couple of days, 00:14:15.500 --> 00:14:19.040 align:middle line:84% you're going to be installing it with Pip Install. 00:14:19.040 --> 00:14:24.590 align:middle line:84% So Music21 really works from the notion that, first, we're 00:14:24.590 --> 00:14:29.060 align:middle line:84% going to encode a lot of ground up concepts about music 00:14:29.060 --> 00:14:31.470 align:middle line:90% so that when you have a chord-- 00:14:31.470 --> 00:14:32.310 align:middle line:90% is that going? 00:14:32.310 --> 00:14:35.420 align:middle line:84% Yep-- we'll just generate a chord, obviously. 00:14:35.420 --> 00:14:36.163 align:middle line:90% Not obviously. 00:14:36.163 --> 00:14:37.830 align:middle line:84% Otherwise, you wouldn't need this class. 00:14:37.830 --> 00:14:40.120 align:middle line:84% But you're going to have a representation of how 00:14:40.120 --> 00:14:43.870 align:middle line:84% to represent pitches, middle C, E, G, and then 00:14:43.870 --> 00:14:46.600 align:middle line:84% all the different things once you have that chord that you 00:14:46.600 --> 00:14:48.940 align:middle line:84% might be able to do, you might get its lyrics out, 00:14:48.940 --> 00:14:51.460 align:middle line:90% and it's going by really fast. 00:14:51.460 --> 00:14:53.000 align:middle line:90% Get its volume. 00:14:53.000 --> 00:14:55.910 align:middle line:90% Is it a diminished triad? 00:14:55.910 --> 00:14:57.110 align:middle line:90% So you can do that. 00:14:57.110 --> 00:15:00.220 align:middle line:84% And hopefully, any good computer programs 00:15:00.220 --> 00:15:03.440 align:middle line:84% seeing C, E, G for is diminished triad will return-- 00:15:03.440 --> 00:15:03.982 align:middle line:90% AUDIENCE: No. 00:15:03.982 --> 00:15:06.107 align:middle line:84% MICHAEL CUTHBERT: No, false, or anything like that. 00:15:06.107 --> 00:15:06.710 align:middle line:90% Zero. 00:15:06.710 --> 00:15:07.515 align:middle line:90% Yep, good. 00:15:07.515 --> 00:15:10.690 align:middle line:90% 00:15:10.690 --> 00:15:13.540 align:middle line:84% So in the amount of time it takes 00:15:13.540 --> 00:15:15.100 align:middle line:84% to write computer program that can 00:15:15.100 --> 00:15:17.950 align:middle line:84% tell whether C, E, G is a diminished triad, 00:15:17.950 --> 00:15:19.760 align:middle line:90% you all can hear it instantly. 00:15:19.760 --> 00:15:22.460 align:middle line:84% So why have a computer do things? 00:15:22.460 --> 00:15:26.350 align:middle line:84% And it's really so that we can listen to music faster than we 00:15:26.350 --> 00:15:28.340 align:middle line:90% can ever listen with our ears. 00:15:28.340 --> 00:15:32.560 align:middle line:84% So this is a particular example of code in Music21, 00:15:32.560 --> 00:15:36.690 align:middle line:84% similar code in other toolkits, that will take a chorale 00:15:36.690 --> 00:15:42.490 align:middle line:84% by Johann Sebastian Bach, his 11th cantata, sixth movement, 00:15:42.490 --> 00:15:44.800 align:middle line:84% encoded as a file called MusicXML, 00:15:44.800 --> 00:15:48.040 align:middle line:84% store it as a variable, get the alto part out of it, 00:15:48.040 --> 00:15:52.620 align:middle line:84% store it as alto, get measures 13 to 20 as excerpt, 00:15:52.620 --> 00:15:55.320 align:middle line:84% do a particular analysis where we're going to put all these 00:15:55.320 --> 00:16:00.420 align:middle line:84% stars on according to how strongly accented something 00:16:00.420 --> 00:16:03.480 align:middle line:84% might be metrically, and show it. 00:16:03.480 --> 00:16:06.210 align:middle line:84% And really, any music analyst, in the amount 00:16:06.210 --> 00:16:09.120 align:middle line:84% of time I took to do this, could have 00:16:09.120 --> 00:16:11.590 align:middle line:84% done it herself, or himself, or themselves. 00:16:11.590 --> 00:16:15.810 align:middle line:84% But what's the great three-letter-- 00:16:15.810 --> 00:16:16.690 align:middle line:90% is it preposition? 00:16:16.690 --> 00:16:21.330 align:middle line:84% Preposition that makes computers really powerful? 00:16:21.330 --> 00:16:24.669 align:middle line:90% Begins with F. 00:16:24.669 --> 00:16:25.252 align:middle line:90% AUDIENCE: For? 00:16:25.252 --> 00:16:26.470 align:middle line:90% MICHAEL CUTHBERT: For, yeah. 00:16:26.470 --> 00:16:30.570 align:middle line:84% So we just put a for loop in, and suddenly, we're 00:16:30.570 --> 00:16:32.580 align:middle line:84% looking at every chorale that Bach 00:16:32.580 --> 00:16:35.240 align:middle line:84% wrote, getting its alto part, measure 13 through 20, 00:16:35.240 --> 00:16:36.480 align:middle line:90% doing the same thing. 00:16:36.480 --> 00:16:39.150 align:middle line:84% And of course, why just do that little excerpt? 00:16:39.150 --> 00:16:40.440 align:middle line:90% Why just do the alto part? 00:16:40.440 --> 00:16:41.750 align:middle line:90% Let's grab every chorale. 00:16:41.750 --> 00:16:43.440 align:middle line:90% For loops within for loops. 00:16:43.440 --> 00:16:45.330 align:middle line:90% Don't over optimize first. 00:16:45.330 --> 00:16:48.350 align:middle line:84% Figure out if you can do this in the time 00:16:48.350 --> 00:16:51.950 align:middle line:84% it takes you to get a cup of coffee. 00:16:51.950 --> 00:16:56.180 align:middle line:84% Yeah, we'll come back to that theme in the class a lot. 00:16:56.180 --> 00:16:58.800 align:middle line:84% In your computer science intro classes, 00:16:58.800 --> 00:17:01.340 align:middle line:84% you're going to early on be thinking a lot about, 00:17:01.340 --> 00:17:04.280 align:middle line:90% is this big O of n squared? 00:17:04.280 --> 00:17:05.430 align:middle line:90% Is this n cubed? 00:17:05.430 --> 00:17:07.829 align:middle line:84% Is this n log n, all these terms? 00:17:07.829 --> 00:17:10.290 align:middle line:84% We're going to talk about that from time to time here. 00:17:10.290 --> 00:17:13.339 align:middle line:84% But for the first part, we're going 00:17:13.339 --> 00:17:17.420 align:middle line:84% to build tools that work in an acceptable, to us, 00:17:17.420 --> 00:17:18.599 align:middle line:90% amount of time. 00:17:18.599 --> 00:17:21.589 align:middle line:84% You're going to hear me use this term a lot when we have a really 00:17:21.589 --> 00:17:25.160 align:middle line:84% big research question about, oh, I don't know, 00:17:25.160 --> 00:17:28.290 align:middle line:90% 15th century musical styles. 00:17:28.290 --> 00:17:33.610 align:middle line:84% And we'll say people have waited 500 years for this answer. 00:17:33.610 --> 00:17:36.280 align:middle line:90% 10 minutes isn't so bad. 00:17:36.280 --> 00:17:39.040 align:middle line:84% And then later, you can go back and make 00:17:39.040 --> 00:17:42.550 align:middle line:84% it a three-minute problem, something like that. 00:17:42.550 --> 00:17:45.850 align:middle line:84% Some of the questions we might ask that come up 00:17:45.850 --> 00:17:49.000 align:middle line:84% that these kinds of tool kits are built to solve 00:17:49.000 --> 00:17:52.480 align:middle line:90% are higher notes louder. 00:17:52.480 --> 00:17:54.230 align:middle line:84% Many of us perceive them as louder. 00:17:54.230 --> 00:17:56.200 align:middle line:90% Are they given louder dynamics? 00:17:56.200 --> 00:17:59.740 align:middle line:90% So we can plot pitch. 00:17:59.740 --> 00:18:02.590 align:middle line:84% If we can go octave numbers, middle C, C4 00:18:02.590 --> 00:18:05.030 align:middle line:84% is the standard American representation today. 00:18:05.030 --> 00:18:08.020 align:middle line:84% And the cello doesn't really go up that high. 00:18:08.020 --> 00:18:08.600 align:middle line:90% I don't know. 00:18:08.600 --> 00:18:11.780 align:middle line:84% Is Sforzando really somewhere between Forte? 00:18:11.780 --> 00:18:12.460 align:middle line:90% Who knows? 00:18:12.460 --> 00:18:14.390 align:middle line:84% These are kind of slightly arbitrary. 00:18:14.390 --> 00:18:17.140 align:middle line:90% But doesn't look like it to me. 00:18:17.140 --> 00:18:19.370 align:middle line:90% Now, this is one little graph. 00:18:19.370 --> 00:18:21.890 align:middle line:84% Probably want to do this on a couple thousand pieces, 00:18:21.890 --> 00:18:24.950 align:middle line:84% run some regression lines, all the things we can do. 00:18:24.950 --> 00:18:27.670 align:middle line:84% But we can start answering those questions. 00:18:27.670 --> 00:18:30.880 align:middle line:84% How does Robert Schumann use notes differently 00:18:30.880 --> 00:18:32.520 align:middle line:90% from Frederic Chopin? 00:18:32.520 --> 00:18:37.990 align:middle line:84% And I plotted this in, plotting the pitch usage of Schumann. 00:18:37.990 --> 00:18:39.940 align:middle line:90% I can see how beautiful. 00:18:39.940 --> 00:18:42.130 align:middle line:90% It's a really nice bell curve. 00:18:42.130 --> 00:18:42.960 align:middle line:90% Really great. 00:18:42.960 --> 00:18:45.190 align:middle line:84% Chopin, on the other hand, looks, 00:18:45.190 --> 00:18:48.570 align:middle line:84% I don't know, kind of like one of these cities with no planning 00:18:48.570 --> 00:18:52.870 align:middle line:84% laws, let's say, Abu Dhabi with a gigantic skyscraper. 00:18:52.870 --> 00:18:56.770 align:middle line:84% When Chopin likes a G-sharp, it's not just any G-sharp. 00:18:56.770 --> 00:18:58.840 align:middle line:90% It's this G-sharp. 00:18:58.840 --> 00:19:01.180 align:middle line:84% They just keep playing it forever. 00:19:01.180 --> 00:19:04.810 align:middle line:84% So there are some of the ways you can do this. 00:19:04.810 --> 00:19:07.900 align:middle line:84% Any Robert Schumann experts out there? 00:19:07.900 --> 00:19:08.400 align:middle line:90% No. 00:19:08.400 --> 00:19:10.030 align:middle line:90% Some hands almost went up. 00:19:10.030 --> 00:19:12.030 align:middle line:84% Some people like Robert Schumann. 00:19:12.030 --> 00:19:14.880 align:middle line:90% Garbage in, garbage out. 00:19:14.880 --> 00:19:18.930 align:middle line:84% This particular piece by Schumann is a string quartet. 00:19:18.930 --> 00:19:20.840 align:middle line:90% This piece by Chopin is a piano. 00:19:20.840 --> 00:19:22.590 align:middle line:84% When you have a string quartet, of course, 00:19:22.590 --> 00:19:24.180 align:middle line:84% you're going to have more instruments 00:19:24.180 --> 00:19:26.160 align:middle line:90% with notes in the middle. 00:19:26.160 --> 00:19:28.750 align:middle line:90% So I did this once. 00:19:28.750 --> 00:19:31.310 align:middle line:84% I thought I had compared apples to apples, 00:19:31.310 --> 00:19:33.630 align:middle line:84% and I realized I compared apples to oranges. 00:19:33.630 --> 00:19:36.440 align:middle line:84% So we're going to think a lot about user error, 00:19:36.440 --> 00:19:40.730 align:middle line:90% also, in how we make things. 00:19:40.730 --> 00:19:44.370 align:middle line:84% Visualization of music is a big part of what we're doing. 00:19:44.370 --> 00:19:46.460 align:middle line:84% These kinds of visualized graphs can 00:19:46.460 --> 00:19:49.490 align:middle line:84% make understanding a piece very easy. 00:19:49.490 --> 00:19:53.218 align:middle line:84% You can also use 3D graphs, which don't make it easy at all 00:19:53.218 --> 00:19:55.760 align:middle line:84% to understand anything, but they look really cool if you need 00:19:55.760 --> 00:19:58.200 align:middle line:84% to donors, and funding, and stuff. 00:19:58.200 --> 00:20:00.660 align:middle line:84% So always make them when you need that. 00:20:00.660 --> 00:20:04.130 align:middle line:84% But otherwise, I can't really read that. 00:20:04.130 --> 00:20:07.250 align:middle line:84% We can combine different fields, so 00:20:07.250 --> 00:20:10.880 align:middle line:84% for instance, geographic information and plotting, 00:20:10.880 --> 00:20:12.660 align:middle line:90% with musical characteristics. 00:20:12.660 --> 00:20:16.010 align:middle line:84% So this is one of my fun things I found. 00:20:16.010 --> 00:20:19.650 align:middle line:84% Comparing the distribution of large intervals, 00:20:19.650 --> 00:20:24.320 align:middle line:84% these are intervals close to what would be Western sevenths. 00:20:24.320 --> 00:20:26.240 align:middle line:84% But obviously, they're not exactly 00:20:26.240 --> 00:20:28.650 align:middle line:84% that because different cultures use different intervals. 00:20:28.650 --> 00:20:31.570 align:middle line:84% But looking at the folk songs differing, 00:20:31.570 --> 00:20:36.550 align:middle line:84% and you can see a large difference as we move northward, 00:20:36.550 --> 00:20:39.230 align:middle line:84% and a little bit as you move inland in China, 00:20:39.230 --> 00:20:41.380 align:middle line:84% in the increase of large intervals 00:20:41.380 --> 00:20:43.430 align:middle line:84% that are not octaves in folk songs. 00:20:43.430 --> 00:20:46.270 align:middle line:84% And we'll look, at a certain point over the semester, 00:20:46.270 --> 00:20:52.420 align:middle line:84% at a very cool paper that kind of shows how particular mountain 00:20:52.420 --> 00:20:54.580 align:middle line:84% range divides folk songs that are 00:20:54.580 --> 00:20:57.735 align:middle line:84% very different on either side of that range. 00:20:57.735 --> 00:21:00.790 align:middle line:90% 00:21:00.790 --> 00:21:04.570 align:middle line:84% If you're looking for particular fingerprints that might tell you 00:21:04.570 --> 00:21:07.000 align:middle line:84% who a composer was, maybe there's 00:21:07.000 --> 00:21:09.200 align:middle line:84% these small moments you might find. 00:21:09.200 --> 00:21:16.150 align:middle line:84% There's one moment in this piece from around 1500 00:21:16.150 --> 00:21:20.350 align:middle line:84% that feels like it might be in a modern composition, 00:21:20.350 --> 00:21:24.700 align:middle line:84% where you just smash your hand down on a piano. 00:21:24.700 --> 00:21:29.010 align:middle line:84% It's very hard to see here, but with a computer, what you can do 00:21:29.010 --> 00:21:33.960 align:middle line:84% is we can take all of the notes and put them into singular 00:21:33.960 --> 00:21:37.810 align:middle line:84% chords here, reduce them all so that they're in closed position. 00:21:37.810 --> 00:21:40.210 align:middle line:84% Everyone remember that from music theory class? 00:21:40.210 --> 00:21:42.010 align:middle line:84% As close together as you can make it. 00:21:42.010 --> 00:21:48.240 align:middle line:84% Annotate the intervals, and suddenly, this one little moment 00:21:48.240 --> 00:21:51.250 align:middle line:90% jumps out as very strange. 00:21:51.250 --> 00:21:53.860 align:middle line:84% Then you can go back and figure out, how did the music, 00:21:53.860 --> 00:21:54.480 align:middle line:90% create this? 00:21:54.480 --> 00:21:57.067 align:middle line:90% 00:21:57.067 --> 00:21:58.650 align:middle line:84% Over the course of the semester, we'll 00:21:58.650 --> 00:22:03.090 align:middle line:84% be looking at changes in music over time. 00:22:03.090 --> 00:22:06.300 align:middle line:84% And we're also going to be looking at various things that 00:22:06.300 --> 00:22:08.430 align:middle line:84% were done by people, either in this class, 00:22:08.430 --> 00:22:13.470 align:middle line:84% or who had worked in the lab that helps create Music21. 00:22:13.470 --> 00:22:17.160 align:middle line:84% So Beth Hadley was somebody who graduated from this program, who 00:22:17.160 --> 00:22:18.180 align:middle line:90% did this analysis. 00:22:18.180 --> 00:22:22.530 align:middle line:84% And looking at a lot of rock popular music songs in the 1950s 00:22:22.530 --> 00:22:27.890 align:middle line:84% and '60s, see the bass moves quite often by five semitones up 00:22:27.890 --> 00:22:31.550 align:middle line:84% or down, which corresponds to a movement of one to four, 00:22:31.550 --> 00:22:36.080 align:middle line:84% or five to one, things like that, and then occasionally, 00:22:36.080 --> 00:22:37.590 align:middle line:90% up by two semitones. 00:22:37.590 --> 00:22:39.960 align:middle line:84% What's that in the 1, 4, 5, 1 progression? 00:22:39.960 --> 00:22:43.290 align:middle line:90% 1, 4, moving to 5, to 1. 00:22:43.290 --> 00:22:45.410 align:middle line:84% And so you can see how this really dominates. 00:22:45.410 --> 00:22:48.890 align:middle line:90% 00:22:48.890 --> 00:22:49.440 align:middle line:90% Did that go? 00:22:49.440 --> 00:22:49.610 align:middle line:90% AUDIENCE: Yeah. 00:22:49.610 --> 00:22:51.965 align:middle line:84% MICHAEL CUTHBERT: At the time, most data set, it 00:22:51.965 --> 00:22:53.060 align:middle line:90% says post 1980. 00:22:53.060 --> 00:22:56.550 align:middle line:84% It should say 1980 to 2000 data set. 00:22:56.550 --> 00:22:58.790 align:middle line:84% And you can see, if we go back and forth, 00:22:58.790 --> 00:23:03.660 align:middle line:84% how it's still these fives and twos that the main thing. 00:23:03.660 --> 00:23:09.830 align:middle line:84% But these other intervals are continuing to move up 00:23:09.830 --> 00:23:12.390 align:middle line:84% and becoming more and more important. 00:23:12.390 --> 00:23:13.800 align:middle line:90% Love to see an analysis. 00:23:13.800 --> 00:23:16.530 align:middle line:84% We now have another full 20 years, 00:23:16.530 --> 00:23:21.380 align:middle line:84% 22.25 years by the time the semester is done, 00:23:21.380 --> 00:23:25.630 align:middle line:84% worth of music after 2000 to see if this trend continues 00:23:25.630 --> 00:23:26.250 align:middle line:90% or stopped. 00:23:26.250 --> 00:23:29.110 align:middle line:90% 00:23:29.110 --> 00:23:32.930 align:middle line:84% Anybody, y'all have taken either some of a theory class, 00:23:32.930 --> 00:23:33.770 align:middle line:90% so at some point. 00:23:33.770 --> 00:23:36.910 align:middle line:84% Who has written a parallel fifth in this class, accidentally? 00:23:36.910 --> 00:23:38.350 align:middle line:90% Yes. 00:23:38.350 --> 00:23:40.790 align:middle line:84% Wow, some people have never written a parallel fifth. 00:23:40.790 --> 00:23:41.630 align:middle line:90% That's amazing. 00:23:41.630 --> 00:23:44.805 align:middle line:84% Or you always did it on purpose just to annoy your teacher. 00:23:44.805 --> 00:23:47.060 align:middle line:90% So next time you. 00:23:47.060 --> 00:23:49.360 align:middle line:84% You've given that you can generate 00:23:49.360 --> 00:23:52.300 align:middle line:84% your own list of every single time 00:23:52.300 --> 00:23:55.900 align:middle line:84% that Johann Sebastian Bach wrote a parallel fifth or an octave, 00:23:55.900 --> 00:23:57.730 align:middle line:84% so that you can give it to your teacher 00:23:57.730 --> 00:24:00.290 align:middle line:84% and say, yeah, that was a parallel fifth. 00:24:00.290 --> 00:24:04.150 align:middle line:84% But it's exactly like what Bach wrote. 00:24:04.150 --> 00:24:06.250 align:middle line:84% Once you've written a program that 00:24:06.250 --> 00:24:08.170 align:middle line:84% does a particular type of analysis 00:24:08.170 --> 00:24:11.630 align:middle line:84% on a particular corpus of music, a collection of music, 00:24:11.630 --> 00:24:14.600 align:middle line:84% an analysis, it's easy to swap one or the other, 00:24:14.600 --> 00:24:16.450 align:middle line:84% so that once you have something that 00:24:16.450 --> 00:24:19.610 align:middle line:84% finds all parallel fifths in Johann Sebastian Bach, 00:24:19.610 --> 00:24:22.350 align:middle line:84% then we can put it into Claudio Monteverdi, 00:24:22.350 --> 00:24:25.000 align:middle line:84% and find the same analysis and do different things. 00:24:25.000 --> 00:24:27.720 align:middle line:84% So each of our tools can work on-- 00:24:27.720 --> 00:24:28.890 align:middle line:90% what is it? 00:24:28.890 --> 00:24:29.838 align:middle line:90% Most of our tools-- 00:24:29.838 --> 00:24:30.880 align:middle line:90% I'll say not all of them. 00:24:30.880 --> 00:24:32.800 align:middle line:84% Most of our tools can work on any corpus, 00:24:32.800 --> 00:24:35.410 align:middle line:84% and any corpus can apply to multiple tools. 00:24:35.410 --> 00:24:40.980 align:middle line:84% So this is a place where you get a squared increase in power 00:24:40.980 --> 00:24:44.720 align:middle line:84% by using the computer over if you wanted to do this by hand. 00:24:44.720 --> 00:24:47.610 align:middle line:90% 00:24:47.610 --> 00:24:50.370 align:middle line:84% I'm trained primarily as a musicologist. 00:24:50.370 --> 00:24:53.190 align:middle line:84% That is a music historian somebody first 00:24:53.190 --> 00:24:56.700 align:middle line:84% trained to do things by hand, and look at history, 00:24:56.700 --> 00:24:59.020 align:middle line:84% and look at context, and come to conclusions. 00:24:59.020 --> 00:25:00.730 align:middle line:84% Then I did a lot of computer things. 00:25:00.730 --> 00:25:04.758 align:middle line:84% And for a little while, these were slightly separate worlds. 00:25:04.758 --> 00:25:06.300 align:middle line:84% I'm trying to move them more and more 00:25:06.300 --> 00:25:09.940 align:middle line:84% together, so that when we answer a particular question, 00:25:09.940 --> 00:25:13.980 align:middle line:84% we think about which of those big questions, what bigger 00:25:13.980 --> 00:25:15.970 align:middle line:90% answer might we have come to. 00:25:15.970 --> 00:25:19.620 align:middle line:84% So this was a particular demonstration program 00:25:19.620 --> 00:25:25.597 align:middle line:84% I wrote early in my development that I was like, oh, cool. 00:25:25.597 --> 00:25:26.430 align:middle line:90% You're going to see. 00:25:26.430 --> 00:25:29.240 align:middle line:84% Well, we'll talk about why JS Bach is coming up a lot. 00:25:29.240 --> 00:25:33.120 align:middle line:84% But I have all these pieces by Bach. 00:25:33.120 --> 00:25:38.160 align:middle line:84% And I know that Bach, when he writes a piece that's in minor-- 00:25:38.160 --> 00:25:41.408 align:middle line:90% [MUSICAL NOTES] 00:25:41.408 --> 00:25:43.730 align:middle line:90% 00:25:43.730 --> 00:25:45.560 align:middle line:84% How is it going to end generally? 00:25:45.560 --> 00:25:48.200 align:middle line:90% Anybody know? 00:25:48.200 --> 00:25:49.740 align:middle line:84% AUDIENCE: And on a C major chord. 00:25:49.740 --> 00:25:51.860 align:middle line:90% [MUSICAL NOTES] 00:25:51.860 --> 00:25:54.660 align:middle line:84% MICHAEL CUTHBERT: Yeah, I call it Picardy third. 00:25:54.660 --> 00:25:58.230 align:middle line:84% So usually, when he's in minor, he ends in major. 00:25:58.230 --> 00:26:01.370 align:middle line:84% I wanted to find every exception to that 00:26:01.370 --> 00:26:03.240 align:middle line:90% in his choral composition. 00:26:03.240 --> 00:26:04.700 align:middle line:90% Great. 00:26:04.700 --> 00:26:09.060 align:middle line:84% So pretty easy to find all these places. 00:26:09.060 --> 00:26:13.640 align:middle line:84% And then I was pretty happy that I found that there's exactly 00:26:13.640 --> 00:26:14.940 align:middle line:90% eight of them. 00:26:14.940 --> 00:26:18.250 align:middle line:84% I thought that was kind of neat, until one day, I 00:26:18.250 --> 00:26:19.750 align:middle line:84% was a little bit bored and I started 00:26:19.750 --> 00:26:21.590 align:middle line:84% looking at what these pieces were, 00:26:21.590 --> 00:26:24.530 align:middle line:84% what pieces went with these weird numbers. 00:26:24.530 --> 00:26:29.900 align:middle line:84% And I looked at the first one, piece 248 in Bach's catalog. 00:26:29.900 --> 00:26:31.655 align:middle line:90% And I saw, oh, that one's wrong. 00:26:31.655 --> 00:26:32.780 align:middle line:90% The computer didn't get it. 00:26:32.780 --> 00:26:34.910 align:middle line:90% I didn't expand all the repeats. 00:26:34.910 --> 00:26:39.475 align:middle line:84% So the final bar in the piece then said, dot capo. 00:26:39.475 --> 00:26:40.850 align:middle line:84% Wait, which way is right for you? 00:26:40.850 --> 00:26:43.100 align:middle line:90% Yeah, da capo al fine. 00:26:43.100 --> 00:26:46.330 align:middle line:84% So go back there, and the actual end was in the middle, 00:26:46.330 --> 00:26:47.750 align:middle line:90% and that one was in major. 00:26:47.750 --> 00:26:49.510 align:middle line:90% So it followed that. 00:26:49.510 --> 00:26:51.620 align:middle line:84% Then I looked at these other pieces, 00:26:51.620 --> 00:26:54.770 align:middle line:84% and I saw they were what some people call orphan corrals. 00:26:54.770 --> 00:26:58.570 align:middle line:84% They're ones that JS Bach's son, CP Bach, 00:26:58.570 --> 00:27:01.010 align:middle line:84% said dad wrote and published them, 00:27:01.010 --> 00:27:03.530 align:middle line:84% but we have no other evidence that dad wrote them. 00:27:03.530 --> 00:27:08.660 align:middle line:84% And the only time this ever happens is in those collections. 00:27:08.660 --> 00:27:10.790 align:middle line:84% So some people have been arguing, 00:27:10.790 --> 00:27:13.000 align:middle line:84% maybe these pieces aren't really by Bach, 00:27:13.000 --> 00:27:15.430 align:middle line:90% or they've been edited heavily. 00:27:15.430 --> 00:27:19.350 align:middle line:84% This seems to give us some of the evidence that might be true. 00:27:19.350 --> 00:27:23.910 align:middle line:84% So always remember that the computer is not 00:27:23.910 --> 00:27:25.510 align:middle line:90% substituting for analysis. 00:27:25.510 --> 00:27:28.380 align:middle line:84% It's a tool to make the kind of analytical questions 00:27:28.380 --> 00:27:32.970 align:middle line:84% you might already have more efficient. 00:27:32.970 --> 00:27:33.670 align:middle line:90% That's why. 00:27:33.670 --> 00:27:34.290 align:middle line:90% Yeah. 00:27:34.290 --> 00:27:37.050 align:middle line:90% That's what we know. 00:27:37.050 --> 00:27:38.910 align:middle line:84% The first couple of things that we're 00:27:38.910 --> 00:27:40.590 align:middle line:84% going to be looking at in this class 00:27:40.590 --> 00:27:45.610 align:middle line:84% are how we represent music, what kinds of scores are happening, 00:27:45.610 --> 00:27:47.370 align:middle line:90% and so on. 00:27:47.370 --> 00:27:49.260 align:middle line:84% And we'll get into why you might do 00:27:49.260 --> 00:27:51.408 align:middle line:90% one representation or another. 00:27:51.408 --> 00:27:52.950 align:middle line:84% One of the first things we'll look at 00:27:52.950 --> 00:27:56.342 align:middle line:84% is music as a series of boxes, of Russian nesting dolls, 00:27:56.342 --> 00:27:57.550 align:middle line:90% whatever you want to call it. 00:27:57.550 --> 00:27:59.110 align:middle line:90% Box, within a box, within a box. 00:27:59.110 --> 00:28:00.510 align:middle line:90% So we have a score. 00:28:00.510 --> 00:28:03.090 align:middle line:84% Inside that box, we have some parts. 00:28:03.090 --> 00:28:05.800 align:middle line:84% Inside those boxes, we might have measures. 00:28:05.800 --> 00:28:08.910 align:middle line:84% We might have voices, two different independent voices 00:28:08.910 --> 00:28:10.870 align:middle line:84% within a measure, something like that. 00:28:10.870 --> 00:28:13.630 align:middle line:84% So we'll call those part based containers. 00:28:13.630 --> 00:28:17.840 align:middle line:84% You might even put the score into a bigger box with the Opus 00:28:17.840 --> 00:28:21.620 align:middle line:84% or all the works in a collection. 00:28:21.620 --> 00:28:27.530 align:middle line:84% And then our second box we might have is everything that happens 00:28:27.530 --> 00:28:29.330 align:middle line:90% at a certain point in time. 00:28:29.330 --> 00:28:32.480 align:middle line:84% So everything that happens here, everything that 00:28:32.480 --> 00:28:34.400 align:middle line:90% happens here, here. 00:28:34.400 --> 00:28:36.410 align:middle line:84% And we'll be looking at and trying 00:28:36.410 --> 00:28:39.860 align:middle line:84% to identify when one representation might be better 00:28:39.860 --> 00:28:42.830 align:middle line:84% and one, another representation might be better, 00:28:42.830 --> 00:28:47.330 align:middle line:84% and how you might use right tools to move from one 00:28:47.330 --> 00:28:52.760 align:middle line:84% representation to another, from a box based to a moment based. 00:28:52.760 --> 00:28:55.850 align:middle line:84% You'll see we sometimes call this salami slicing. 00:28:55.850 --> 00:28:58.000 align:middle line:84% You just chop the thing like a salami. 00:28:58.000 --> 00:29:01.580 align:middle line:90% 00:29:01.580 --> 00:29:06.600 align:middle line:84% A really important moment, and I alluded to with Bach there, 00:29:06.600 --> 00:29:11.690 align:middle line:84% is to try to make sure that the conclusions that we're making 00:29:11.690 --> 00:29:14.700 align:middle line:84% apply to more music that's out there. 00:29:14.700 --> 00:29:18.130 align:middle line:84% And that's only possible if we have a lot of music, 00:29:18.130 --> 00:29:22.950 align:middle line:84% and that music is representative of people who create music. 00:29:22.950 --> 00:29:27.040 align:middle line:84% Here's some of my favorite composers out there. 00:29:27.040 --> 00:29:28.480 align:middle line:90% We have Ludwig van Beethoven. 00:29:28.480 --> 00:29:32.380 align:middle line:84% We have Joseph Bologne, Chevalier de Saint-Georges, 00:29:32.380 --> 00:29:34.630 align:middle line:84% Kate Soper, friend of this music department, 00:29:34.630 --> 00:29:40.480 align:middle line:84% Amy Beach, Wagner, Richard Wagner, JS Bach, Scott Joplin, 00:29:40.480 --> 00:29:42.010 align:middle line:90% Queen Lili'uokalani. 00:29:42.010 --> 00:29:43.540 align:middle line:84% I spend part of my year in Hawaii. 00:29:43.540 --> 00:29:46.440 align:middle line:90% Very important composer. 00:29:46.440 --> 00:29:46.990 align:middle line:90% Let me see. 00:29:46.990 --> 00:29:50.430 align:middle line:90% Yamandu Costa, Wolfgang Mozart. 00:29:50.430 --> 00:29:52.300 align:middle line:84% We've got Dolly Parton and the Beatles. 00:29:52.300 --> 00:29:57.360 align:middle line:84% So these are people that I would really love to do my work on. 00:29:57.360 --> 00:29:59.620 align:middle line:84% So I go and I start doing the work. 00:29:59.620 --> 00:30:04.140 align:middle line:84% And of course, I don't want to encode all of this music 00:30:04.140 --> 00:30:07.050 align:middle line:84% in, so I go with what other people have encoded. 00:30:07.050 --> 00:30:11.250 align:middle line:84% And that lets me work on this subset of music. 00:30:11.250 --> 00:30:14.300 align:middle line:84% Very representative subset of the people 00:30:14.300 --> 00:30:17.240 align:middle line:90% who were up there before? 00:30:17.240 --> 00:30:18.690 align:middle line:90% No, not at all. 00:30:18.690 --> 00:30:26.160 align:middle line:84% We have Beethoven, Bach, Wagner, Mozart, all great composers, 00:30:26.160 --> 00:30:29.370 align:middle line:84% long dead white males, out of copyright. 00:30:29.370 --> 00:30:31.910 align:middle line:84% And then still in copyright are white 00:30:31.910 --> 00:30:34.340 align:middle line:84% male group, The Beatles, that somehow, there's 00:30:34.340 --> 00:30:36.320 align:middle line:90% just always Beatles everywhere. 00:30:36.320 --> 00:30:37.890 align:middle line:90% You can almost always get that. 00:30:37.890 --> 00:30:42.740 align:middle line:84% So this was the state of music, which sometime 00:30:42.740 --> 00:30:45.770 align:middle line:84% is called corpus research, work on large collections of scores, 00:30:45.770 --> 00:30:49.610 align:middle line:84% a corpus body, a large body, a repertoire of scores 00:30:49.610 --> 00:30:53.060 align:middle line:90% when I started going into this. 00:30:53.060 --> 00:30:59.040 align:middle line:84% Thankfully, the situation is getting a little, little better. 00:30:59.040 --> 00:31:01.010 align:middle line:84% We now have the complete piano rags 00:31:01.010 --> 00:31:04.170 align:middle line:84% of Scott Joplin, who wrote The Entertainer, Maple Leaf Rag. 00:31:04.170 --> 00:31:06.385 align:middle line:84% If you don't think you know his name, play some. 00:31:06.385 --> 00:31:08.760 align:middle line:84% We'll play some of his music, and you'll think, oh, yeah, 00:31:08.760 --> 00:31:09.720 align:middle line:90% I know that music. 00:31:09.720 --> 00:31:12.620 align:middle line:84% So very important composer we're starting to get in. 00:31:12.620 --> 00:31:16.180 align:middle line:84% And individuals and groups are starting 00:31:16.180 --> 00:31:20.530 align:middle line:84% to create collections, like this collection of a couple hundred 00:31:20.530 --> 00:31:24.640 align:middle line:84% scores by women and nonwhite composers. 00:31:24.640 --> 00:31:29.350 align:middle line:84% I hope that this class, at some point, we will all 00:31:29.350 --> 00:31:34.750 align:middle line:84% contribute onto improving the situation just a bit 00:31:34.750 --> 00:31:40.000 align:middle line:84% and try to make it so, eventually, what is represented 00:31:40.000 --> 00:31:44.265 align:middle line:84% is in our representations looks like the world we want to be in. 00:31:44.265 --> 00:31:47.230 align:middle line:90% 00:31:47.230 --> 00:31:53.830 align:middle line:84% So you have in front of you somewhere a syllabus, as well. 00:31:53.830 --> 00:31:55.460 align:middle line:90% So grab the syllabus. 00:31:55.460 --> 00:31:57.500 align:middle line:84% We're not going to go through it line by line. 00:31:57.500 --> 00:32:00.168 align:middle line:84% In fact, I haven't even given you yet 00:32:00.168 --> 00:32:01.960 align:middle line:84% the longer one that's going to tell you day 00:32:01.960 --> 00:32:03.210 align:middle line:90% by day what we're going to do. 00:32:03.210 --> 00:32:06.520 align:middle line:84% Because part of that depended on how many of y'all showed up 00:32:06.520 --> 00:32:07.100 align:middle line:90% today. 00:32:07.100 --> 00:32:09.050 align:middle line:90% And it's a good, good crowd. 00:32:09.050 --> 00:32:11.130 align:middle line:90% Everybody comfortable? 00:32:11.130 --> 00:32:12.630 align:middle line:90% A little bit, a little bit not. 00:32:12.630 --> 00:32:16.725 align:middle line:84% It's over there, but we'll somehow make this happen. 00:32:16.725 --> 00:32:19.290 align:middle line:90% 00:32:19.290 --> 00:32:23.140 align:middle line:84% You'll be happy that this class, there's nothing to buy, 00:32:23.140 --> 00:32:26.320 align:middle line:84% unless you don't happen to have some of these things earlier. 00:32:26.320 --> 00:32:30.940 align:middle line:84% We'll use as a textbook the Music21, the standard textbook, 00:32:30.940 --> 00:32:33.630 align:middle line:90% the Music21 user's guide. 00:32:33.630 --> 00:32:36.310 align:middle line:84% Probably get to maybe Chapter 25. 00:32:36.310 --> 00:32:39.630 align:middle line:84% I always say 1 to 30, but we might not get there. 00:32:39.630 --> 00:32:41.572 align:middle line:84% And we can't go too much past 30, 00:32:41.572 --> 00:32:43.030 align:middle line:84% because it just kind of trails off. 00:32:43.030 --> 00:32:44.170 align:middle line:90% It's still being written. 00:32:44.170 --> 00:32:45.810 align:middle line:84% You can always write some more of it 00:32:45.810 --> 00:32:49.440 align:middle line:84% if you want as part of a project. 00:32:49.440 --> 00:32:53.260 align:middle line:84% You're required for me to support you, 00:32:53.260 --> 00:32:57.490 align:middle line:84% for me to be a help to have a Windows or Mac OS laptop. 00:32:57.490 --> 00:33:00.490 align:middle line:84% We're going to install the latest version of Music21, 00:33:00.490 --> 00:33:02.970 align:middle line:90% which is, I believe, 8.1. 00:33:02.970 --> 00:33:07.140 align:middle line:84% The hope, the plan, the bags under my eyes 00:33:07.140 --> 00:33:11.210 align:middle line:84% right now was to get version 9 released before today. 00:33:11.210 --> 00:33:12.480 align:middle line:90% It's not out. 00:33:12.480 --> 00:33:16.790 align:middle line:84% It has some of the newer features that we'll be using. 00:33:16.790 --> 00:33:19.230 align:middle line:84% I'll probably get that out within a week or two. 00:33:19.230 --> 00:33:22.770 align:middle line:84% Now, you all are my focus, and then we'll put that on. 00:33:22.770 --> 00:33:26.270 align:middle line:84% So at some point, we'll all upgrade to that. 00:33:26.270 --> 00:33:29.610 align:middle line:84% Music21 version 8 runs on Python 3.8. 00:33:29.610 --> 00:33:34.950 align:middle line:84% Version 9 runs on 3.10, so install of Python 3.10. 00:33:34.950 --> 00:33:38.360 align:middle line:84% You'll be very happy with that or 3.11. 00:33:38.360 --> 00:33:40.820 align:middle line:84% Probably want to get some staff paper and a pencil, 00:33:40.820 --> 00:33:44.040 align:middle line:84% because at various points, unless if you're using a tablet, 00:33:44.040 --> 00:33:44.970 align:middle line:90% you're probably fine. 00:33:44.970 --> 00:33:48.200 align:middle line:84% But it's sometimes hard if you're taking notes on laptop 00:33:48.200 --> 00:33:51.958 align:middle line:84% and we'll, at some point, be using that staff a lot. 00:33:51.958 --> 00:33:53.250 align:middle line:90% There's four recommended books. 00:33:53.250 --> 00:33:55.140 align:middle line:84% I love them, but they're all ebooks, 00:33:55.140 --> 00:33:57.560 align:middle line:84% but you might consider buying them. 00:33:57.560 --> 00:34:00.030 align:middle line:84% At least the first two are pretty cheap and things, 00:34:00.030 --> 00:34:03.320 align:middle line:84% but we don't really need to talk about that more. 00:34:03.320 --> 00:34:06.630 align:middle line:90% OK, we're making great time. 00:34:06.630 --> 00:34:12.699 align:middle line:84% The typical week, we're going to, on Wednesday, introduce 00:34:12.699 --> 00:34:13.880 align:middle line:90% a particularly new topic. 00:34:13.880 --> 00:34:15.987 align:middle line:84% This is just middle of the semester, 00:34:15.987 --> 00:34:18.320 align:middle line:84% And then you're going to get some readings to do for it. 00:34:18.320 --> 00:34:20.300 align:middle line:84% Some of them will probably be for Friday. 00:34:20.300 --> 00:34:25.030 align:middle line:84% We'll try to make Wednesday to Friday kind of a time more 00:34:25.030 --> 00:34:26.949 align:middle line:84% for reading, and reflection and things. 00:34:26.949 --> 00:34:29.980 align:middle line:84% On Friday, so maybe the topic is-- 00:34:29.980 --> 00:34:32.120 align:middle line:90% oh, let me not be hypothetical. 00:34:32.120 --> 00:34:33.110 align:middle line:90% What is Topic 3? 00:34:33.110 --> 00:34:36.159 align:middle line:90% It's written there somewhere. 00:34:36.159 --> 00:34:36.860 align:middle line:90% Containers. 00:34:36.860 --> 00:34:40.159 align:middle line:84% We just talked about this, how to represent as containers. 00:34:40.159 --> 00:34:42.320 align:middle line:84% What are the various ways working through that? 00:34:42.320 --> 00:34:46.210 align:middle line:84% So we'll just introduce, like we did now, a new topic 00:34:46.210 --> 00:34:47.840 align:middle line:90% and think about it a bit. 00:34:47.840 --> 00:34:51.520 align:middle line:84% On Friday, we'll get to some of the nitty gritty, difficult 00:34:51.520 --> 00:34:55.280 align:middle line:84% things, discussion of why this representation work, 00:34:55.280 --> 00:34:57.800 align:middle line:84% why this doesn't, why you might make this choice. 00:34:57.800 --> 00:35:00.380 align:middle line:84% And we'll do some in class work on that. 00:35:00.380 --> 00:35:03.380 align:middle line:84% And then usually, on Fridays, you'll get a problem set. 00:35:03.380 --> 00:35:05.770 align:middle line:84% So you'll get that on work on it. 00:35:05.770 --> 00:35:10.950 align:middle line:84% But over the weekend you'll also be getting a little bit 00:35:10.950 --> 00:35:12.850 align:middle line:90% of asynchronous content. 00:35:12.850 --> 00:35:14.703 align:middle line:90% So what happened was-- 00:35:14.703 --> 00:35:16.870 align:middle line:84% nobody really cares about the history of your class, 00:35:16.870 --> 00:35:18.390 align:middle line:84% but I'm going to tell it anyway-- 00:35:18.390 --> 00:35:19.510 align:middle line:90% we started this class. 00:35:19.510 --> 00:35:21.670 align:middle line:84% It was an hour and a half, twice a week. 00:35:21.670 --> 00:35:25.170 align:middle line:84% And then COVID hit, and suddenly, a lot of things 00:35:25.170 --> 00:35:29.130 align:middle line:84% went online and I did a lot of video recording, partially 00:35:29.130 --> 00:35:30.630 align:middle line:84% because my wife and I were teaching 00:35:30.630 --> 00:35:34.170 align:middle line:84% at the exact same time in a one-bedroom house that had 00:35:34.170 --> 00:35:36.700 align:middle line:90% no way for us to both teach. 00:35:36.700 --> 00:35:38.340 align:middle line:84% So we did some asynchronous things 00:35:38.340 --> 00:35:40.840 align:middle line:84% and a lot of the asynchronous content 00:35:40.840 --> 00:35:43.590 align:middle line:84% I'm so glad you will never have to see again. 00:35:43.590 --> 00:35:45.940 align:middle line:90% So you all lived through that. 00:35:45.940 --> 00:35:49.330 align:middle line:84% But some of it was actually better than in class, 00:35:49.330 --> 00:35:52.210 align:middle line:84% because sometimes, every once in a while, 00:35:52.210 --> 00:35:54.570 align:middle line:84% there's something that I can teach you 00:35:54.570 --> 00:35:56.770 align:middle line:84% a particular technique, and then if it was in class, 00:35:56.770 --> 00:35:59.110 align:middle line:84% I'd say, great, so now you know how to do that, and let's go on. 00:35:59.110 --> 00:36:01.027 align:middle line:84% And here's another technique, and let's go on. 00:36:01.027 --> 00:36:04.000 align:middle line:84% And you haven't had any chance to practice any of them. 00:36:04.000 --> 00:36:06.650 align:middle line:84% So there's some things that are useful about the Sesame Street 00:36:06.650 --> 00:36:10.010 align:middle line:84% style of here's a particular thing that takes me 00:36:10.010 --> 00:36:12.440 align:middle line:84% about five minutes to present, and then you work on it 00:36:12.440 --> 00:36:13.260 align:middle line:90% for 10 minutes. 00:36:13.260 --> 00:36:14.640 align:middle line:84% Then here's another five minutes. 00:36:14.640 --> 00:36:15.390 align:middle line:90% You work on it 10. 00:36:15.390 --> 00:36:17.660 align:middle line:84% Before you know it, you've had one hour of lecture 00:36:17.660 --> 00:36:20.030 align:middle line:84% and you've had about three hours of a problem set 00:36:20.030 --> 00:36:22.400 align:middle line:84% that you've done problem by problem. 00:36:22.400 --> 00:36:23.760 align:middle line:90% So those ones are still there. 00:36:23.760 --> 00:36:27.300 align:middle line:84% There's about seven of times during the semester. 00:36:27.300 --> 00:36:29.990 align:middle line:84% And I just guarantee I wouldn't be 00:36:29.990 --> 00:36:32.780 align:middle line:84% doing that if I didn't think that was a better way 00:36:32.780 --> 00:36:33.330 align:middle line:90% to do things. 00:36:33.330 --> 00:36:37.200 align:middle line:84% So you'll do some of that work there on Wednesday, 00:36:37.200 --> 00:36:41.040 align:middle line:84% and this will be at least four or five hours before class, 00:36:41.040 --> 00:36:43.640 align:middle line:84% so I can do a quick spot check-- that's the great part about 00:36:43.640 --> 00:36:46.170 align:middle line:90% doing things all online-- 00:36:46.170 --> 00:36:52.770 align:middle line:84% turning the problem set, and then we'll unlock the tools. 00:36:52.770 --> 00:36:53.773 align:middle line:90% What does that mean? 00:36:53.773 --> 00:36:55.440 align:middle line:84% It means if you're working on something, 00:36:55.440 --> 00:36:58.050 align:middle line:84% let's say we're doing something with cords. 00:36:58.050 --> 00:37:00.500 align:middle line:84% I'm going to ask you to figure out 00:37:00.500 --> 00:37:02.410 align:middle line:84% certain of the algorithms for how 00:37:02.410 --> 00:37:05.110 align:middle line:84% to solve some difficult problems with chords. 00:37:05.110 --> 00:37:07.870 align:middle line:84% And most of the things that you're going to be doing, 00:37:07.870 --> 00:37:09.740 align:middle line:84% somebody in the world has solved. 00:37:09.740 --> 00:37:11.630 align:middle line:84% I hope that's not too disappointing. 00:37:11.630 --> 00:37:13.342 align:middle line:84% Most of what we do in most of our classes 00:37:13.342 --> 00:37:15.800 align:middle line:84% in structural engineering is saying somebody has solved it. 00:37:15.800 --> 00:37:18.730 align:middle line:84% We're learning to do it ourselves. 00:37:18.730 --> 00:37:21.190 align:middle line:84% So at that point, we're going to show you 00:37:21.190 --> 00:37:23.840 align:middle line:84% some of the tools that are already out there to do this, 00:37:23.840 --> 00:37:25.760 align:middle line:84% and they will become unlocked to you. 00:37:25.760 --> 00:37:28.180 align:middle line:84% That is to say, you can keep doing, using the tools 00:37:28.180 --> 00:37:33.220 align:middle line:84% that you built, or you can use the standardized chord 00:37:33.220 --> 00:37:35.200 align:middle line:90% manipulation tools. 00:37:35.200 --> 00:37:37.000 align:middle line:84% And then we'll introduce the next topic 00:37:37.000 --> 00:37:38.463 align:middle line:90% and keep going over here. 00:37:38.463 --> 00:37:39.880 align:middle line:84% It does mean that you're sometimes 00:37:39.880 --> 00:37:42.397 align:middle line:84% going to have two topics going simultaneously. 00:37:42.397 --> 00:37:43.730 align:middle line:90% And I hope that's not too weird. 00:37:43.730 --> 00:37:46.990 align:middle line:84% I'm going to try to make that as unconfusing as possible. 00:37:46.990 --> 00:37:54.080 align:middle line:84% So if you could get out your laptop and click on this button, 00:37:54.080 --> 00:37:56.110 align:middle line:90% I'll click on it with you. 00:37:56.110 --> 00:37:58.540 align:middle line:84% I'm just going to get this ready to go. 00:37:58.540 --> 00:38:06.970 align:middle line:84% When it all loads, go to Runtime, Run All, and then stop. 00:38:06.970 --> 00:38:09.810 align:middle line:84% That will probably take two to three minutes. 00:38:09.810 --> 00:38:13.140 align:middle line:84% OK, so we'll go on from that, and I'll just 00:38:13.140 --> 00:38:16.330 align:middle line:84% talk a little bit about final projects while that's going. 00:38:16.330 --> 00:38:16.985 align:middle line:90% There we are. 00:38:16.985 --> 00:38:17.485 align:middle line:90% Great. 00:38:17.485 --> 00:38:20.160 align:middle line:90% 00:38:20.160 --> 00:38:22.560 align:middle line:84% So what we're going to do at the end of the semester 00:38:22.560 --> 00:38:23.590 align:middle line:90% is have final projects. 00:38:23.590 --> 00:38:25.030 align:middle line:90% We're a little bit big class. 00:38:25.030 --> 00:38:26.520 align:middle line:84% I'm probably going to let anybody 00:38:26.520 --> 00:38:28.890 align:middle line:84% who wants to have a group of three 00:38:28.890 --> 00:38:31.450 align:middle line:84% do that because of the number of people here. 00:38:31.450 --> 00:38:35.130 align:middle line:84% But generally, two works really, really well. 00:38:35.130 --> 00:38:37.110 align:middle line:84% Solo groups are really hard, and I 00:38:37.110 --> 00:38:39.210 align:middle line:84% found that groups bigger than three 00:38:39.210 --> 00:38:42.070 align:middle line:84% have almost always had some problems. 00:38:42.070 --> 00:38:44.040 align:middle line:84% So we're going to do a final project. 00:38:44.040 --> 00:38:47.640 align:middle line:84% And I used to say any topic in computational music 00:38:47.640 --> 00:38:51.607 align:middle line:84% or music theory and analysis, or composition type things, 00:38:51.607 --> 00:38:53.190 align:middle line:84% as long as it involves representations 00:38:53.190 --> 00:38:54.490 align:middle line:90% of scores for the computer. 00:38:54.490 --> 00:38:56.874 align:middle line:90% I now have one exception. 00:38:56.874 --> 00:39:02.540 align:middle line:84% A lot of AI deep learning for composition projects have tended 00:39:02.540 --> 00:39:07.370 align:middle line:84% not to do as well as the students who ran them, 00:39:07.370 --> 00:39:08.490 align:middle line:90% were doing in class. 00:39:08.490 --> 00:39:12.020 align:middle line:84% So I'm going to not allow that for this semester 00:39:12.020 --> 00:39:15.420 align:middle line:84% to try to figure out how we can do that a little bit better. 00:39:15.420 --> 00:39:19.610 align:middle line:84% So that's just me learning from experience. 00:39:19.610 --> 00:39:23.120 align:middle line:84% You're going to spend some time researching a topic that you 00:39:23.120 --> 00:39:26.220 align:middle line:84% work on, programming the tools needed to do it, 00:39:26.220 --> 00:39:28.875 align:middle line:84% reading about what mistakes you might have done 00:39:28.875 --> 00:39:30.000 align:middle line:90% and just thinking about it. 00:39:30.000 --> 00:39:32.270 align:middle line:84% Then what I want from you is, then you're 00:39:32.270 --> 00:39:36.950 align:middle line:84% going to be making a video about your project, five minutes. 00:39:36.950 --> 00:39:38.187 align:middle line:90% I know I'm a hard professor. 00:39:38.187 --> 00:39:39.770 align:middle line:84% I could have made it 10 minutes and it 00:39:39.770 --> 00:39:40.978 align:middle line:90% would have been a lot easier. 00:39:40.978 --> 00:39:44.040 align:middle line:84% But five minutes, you have to be on your toes. 00:39:44.040 --> 00:39:47.480 align:middle line:84% So what I want a couple of days before that is just 00:39:47.480 --> 00:39:49.790 align:middle line:90% a draft of your script. 00:39:49.790 --> 00:39:51.240 align:middle line:90% I don't care how messy it is. 00:39:51.240 --> 00:39:54.350 align:middle line:84% I just want to see that I'm not going to get a frantic phone 00:39:54.350 --> 00:39:58.300 align:middle line:84% call at 11:30 PM that none of our software works or something 00:39:58.300 --> 00:39:59.090 align:middle line:90% like that. 00:39:59.090 --> 00:40:03.520 align:middle line:84% And then you record and edit a five minute video. 00:40:03.520 --> 00:40:06.940 align:middle line:84% So let's get back to that live code. 00:40:06.940 --> 00:40:11.180 align:middle line:84% And somewhere on here, this one, we're all up. 00:40:11.180 --> 00:40:14.590 align:middle line:90% And where do I have it? 00:40:14.590 --> 00:40:17.320 align:middle line:90% I have mine here. 00:40:17.320 --> 00:40:20.400 align:middle line:84% Let me make this bigger in three ways. 00:40:20.400 --> 00:40:24.250 align:middle line:90% 00:40:24.250 --> 00:40:27.680 align:middle line:90% Anybody not get a note here? 00:40:27.680 --> 00:40:30.580 align:middle line:84% And maybe this one sometimes doesn't 00:40:30.580 --> 00:40:32.180 align:middle line:90% work for a lot of people. 00:40:32.180 --> 00:40:34.250 align:middle line:84% So don't worry if that doesn't work for you. 00:40:34.250 --> 00:40:37.860 align:middle line:90% But give me a thumbs up. 00:40:37.860 --> 00:40:38.360 align:middle line:90% We got it? 00:40:38.360 --> 00:40:39.050 align:middle line:90% OK, cool. 00:40:39.050 --> 00:40:45.280 align:middle line:84% So let's start with me finding my script. 00:40:45.280 --> 00:40:45.790 align:middle line:90% Great. 00:40:45.790 --> 00:40:50.410 align:middle line:84% So let's do some of the things we can do. 00:40:50.410 --> 00:40:53.510 align:middle line:84% If you don't care about polluting your code base-- 00:40:53.510 --> 00:40:55.390 align:middle line:90% so I got this a little bit-- 00:40:55.390 --> 00:40:58.420 align:middle line:84% go ahead and import from Music21. 00:40:58.420 --> 00:40:59.050 align:middle line:90% Import Star. 00:40:59.050 --> 00:41:01.560 align:middle line:90% It should already be in there. 00:41:01.560 --> 00:41:05.920 align:middle line:84% Has anybody not used Jupyter Notebook or anything like that? 00:41:05.920 --> 00:41:08.020 align:middle line:90% Who has used one? 00:41:08.020 --> 00:41:10.350 align:middle line:84% So don't need to do too much but we 00:41:10.350 --> 00:41:13.810 align:middle line:84% can-- yeah, we'll enter in a new code thing, move it down. 00:41:13.810 --> 00:41:16.510 align:middle line:84% So go ahead, enter a new code bank, 00:41:16.510 --> 00:41:18.180 align:middle line:84% and let's just make sure that if I 00:41:18.180 --> 00:41:24.150 align:middle line:84% type chord and hit Shift, Enter, something like that comes out. 00:41:24.150 --> 00:41:25.240 align:middle line:90% Cool. 00:41:25.240 --> 00:41:25.990 align:middle line:90% Great. 00:41:25.990 --> 00:41:30.330 align:middle line:84% So there's a particular piece by Bach 00:41:30.330 --> 00:41:32.260 align:middle line:84% that we'll use quite often in this class. 00:41:32.260 --> 00:41:35.400 align:middle line:84% I'm going to use it less often this year because we have more 00:41:35.400 --> 00:41:38.460 align:middle line:84% alternatives now to this, and we don't want to perpetuate 00:41:38.460 --> 00:41:40.330 align:middle line:84% the "Bach is everything" stereotype. 00:41:40.330 --> 00:41:42.130 align:middle line:90% But this one we'll use a lot. 00:41:42.130 --> 00:41:44.610 align:middle line:84% So we'll assign to the variable Bach something 00:41:44.610 --> 00:41:50.710 align:middle line:84% from the Music21 corpus, calling the method parse on it. 00:41:50.710 --> 00:41:58.130 align:middle line:90% And we can just type BWV 66.6. 00:41:58.130 --> 00:42:00.990 align:middle line:84% I am not a devil worshipper or something. 00:42:00.990 --> 00:42:05.760 align:middle line:84% 66.6 happens to be the shortest chorale that Bach ever wrote, 00:42:05.760 --> 00:42:09.350 align:middle line:84% so it's a nice one for us to work with here. 00:42:09.350 --> 00:42:14.120 align:middle line:84% And Shift, Enter, and that should load. 00:42:14.120 --> 00:42:17.263 align:middle line:84% If something doesn't work, this is the time 00:42:17.263 --> 00:42:18.680 align:middle line:84% to cheat and look up your neighbor 00:42:18.680 --> 00:42:21.990 align:middle line:84% and see if maybe you mistyped something or tap them on. 00:42:21.990 --> 00:42:23.550 align:middle line:90% We're all in class. 00:42:23.550 --> 00:42:26.490 align:middle line:84% We're all here always just to help each other, right? 00:42:26.490 --> 00:42:28.100 align:middle line:90% We feel that? 00:42:28.100 --> 00:42:30.120 align:middle line:84% You're going to do that for people around you. 00:42:30.120 --> 00:42:31.230 align:middle line:90% Good. 00:42:31.230 --> 00:42:33.390 align:middle line:90% So let's show it. 00:42:33.390 --> 00:42:34.850 align:middle line:84% The first time you run show, it's 00:42:34.850 --> 00:42:37.470 align:middle line:84% always a lot slower than every other time. 00:42:37.470 --> 00:42:39.120 align:middle line:84% Hopefully, it won't take too long. 00:42:39.120 --> 00:42:43.640 align:middle line:84% You should get a score up there somewhere. 00:42:43.640 --> 00:42:44.390 align:middle line:90% Good. 00:42:44.390 --> 00:42:47.840 align:middle line:84% Right now, you're only going to get the first page of the score. 00:42:47.840 --> 00:42:51.350 align:middle line:84% So that's why I use this one, because it's only one page long. 00:42:51.350 --> 00:42:55.250 align:middle line:90% Soon you'll get the whole score. 00:42:55.250 --> 00:43:00.020 align:middle line:84% So let's go ahead and let's get that alto part. 00:43:00.020 --> 00:43:02.980 align:middle line:90% So we'll say alto equals Bach. 00:43:02.980 --> 00:43:07.120 align:middle line:84% Parts is a way of getting something out alto, 00:43:07.120 --> 00:43:09.235 align:middle line:90% and then alto dot show. 00:43:09.235 --> 00:43:17.650 align:middle line:90% 00:43:17.650 --> 00:43:19.150 align:middle line:90% Everybody see something? 00:43:19.150 --> 00:43:22.010 align:middle line:90% It looks a bit like this. 00:43:22.010 --> 00:43:23.560 align:middle line:90% Alto part's always a little-- 00:43:23.560 --> 00:43:24.970 align:middle line:90% [MUSICAL NOTES] 00:43:24.970 --> 00:43:28.260 align:middle line:90% [SINGING] 00:43:28.260 --> 00:43:30.820 align:middle line:90% 00:43:30.820 --> 00:43:32.950 align:middle line:90% Something like that. 00:43:32.950 --> 00:43:34.420 align:middle line:90% Good. 00:43:34.420 --> 00:43:38.420 align:middle line:84% You can also get a part by doing it numerically. 00:43:38.420 --> 00:43:42.070 align:middle line:84% So anybody know in the course what are the four 00:43:42.070 --> 00:43:43.880 align:middle line:90% parts from highest to lowest? 00:43:43.880 --> 00:43:44.630 align:middle line:90% AUDIENCE: Soprano. 00:43:44.630 --> 00:43:45.200 align:middle line:90% MICHAEL CUTHBERT: Soprano? 00:43:45.200 --> 00:43:45.575 align:middle line:90% AUDIENCE: Alto? 00:43:45.575 --> 00:43:45.700 align:middle line:90% AUDIENCE: Alto? 00:43:45.700 --> 00:43:46.870 align:middle line:90% MICHAEL CUTHBERT: Alto? 00:43:46.870 --> 00:43:47.380 align:middle line:90% Tenor? 00:43:47.380 --> 00:43:48.110 align:middle line:90% Louder, louder. 00:43:48.110 --> 00:43:48.380 align:middle line:90% Come on. 00:43:48.380 --> 00:43:49.250 align:middle line:90% Participation grade. 00:43:49.250 --> 00:43:50.540 align:middle line:90% Let's already get our thing. 00:43:50.540 --> 00:43:53.830 align:middle line:90% Soprano, alto, tenor, bass. 00:43:53.830 --> 00:43:54.330 align:middle line:90% Great. 00:43:54.330 --> 00:43:57.390 align:middle line:84% So if we can do it by numeric order, 00:43:57.390 --> 00:44:02.554 align:middle line:84% if we want to get the alto, that's what number part? 00:44:02.554 --> 00:44:03.520 align:middle line:90% Two, one. 00:44:03.520 --> 00:44:05.380 align:middle line:90% Yes, zero index Python. 00:44:05.380 --> 00:44:07.630 align:middle line:84% So what are the things you'll get, 00:44:07.630 --> 00:44:10.230 align:middle line:84% and so that should be the same part, parts one, good. 00:44:10.230 --> 00:44:10.998 align:middle line:90% Good guesses. 00:44:10.998 --> 00:44:11.790 align:middle line:90% Good wrong guesses. 00:44:11.790 --> 00:44:16.110 align:middle line:84% I love great wrong guesses better than even getting 00:44:16.110 --> 00:44:17.260 align:middle line:90% it right on the first time. 00:44:17.260 --> 00:44:19.020 align:middle line:84% One of the things you'll find is, 00:44:19.020 --> 00:44:21.850 align:middle line:84% if something is in computer speak, we leave it alone. 00:44:21.850 --> 00:44:25.140 align:middle line:84% So we don't tend to number the parts in the orchestra. 00:44:25.140 --> 00:44:26.950 align:middle line:90% Hey, instrument 17, do this. 00:44:26.950 --> 00:44:27.450 align:middle line:90% No. 00:44:27.450 --> 00:44:29.370 align:middle line:84% So when we do that, we leave it in computer 00:44:29.370 --> 00:44:34.600 align:middle line:90% speak and make it zero indexed. 00:44:34.600 --> 00:44:39.280 align:middle line:84% When we leave things that musicians might know, 00:44:39.280 --> 00:44:44.800 align:middle line:84% however, we tend to use the musicians thing. 00:44:44.800 --> 00:44:48.630 align:middle line:84% So let's get the soprano part, part zero here. 00:44:48.630 --> 00:44:54.330 align:middle line:84% And let's get measures four through six, 00:44:54.330 --> 00:44:57.900 align:middle line:84% so dot measures 4, comma 6 dot show. 00:44:57.900 --> 00:45:02.420 align:middle line:84% And in Python, that should get you the fifth measure 00:45:02.420 --> 00:45:06.980 align:middle line:84% through the sixth measure because of the way zero indexed. 00:45:06.980 --> 00:45:08.390 align:middle line:90% However, measure numbers. 00:45:08.390 --> 00:45:12.620 align:middle line:84% Measure numbers something that human musicians talk about? 00:45:12.620 --> 00:45:13.310 align:middle line:90% Yeah? 00:45:13.310 --> 00:45:13.860 align:middle line:90% Yeah. 00:45:13.860 --> 00:45:15.000 align:middle line:90% You've heard of that? 00:45:15.000 --> 00:45:15.500 align:middle line:90% Great. 00:45:15.500 --> 00:45:18.200 align:middle line:84% We'll save time always by just being 00:45:18.200 --> 00:45:20.810 align:middle line:84% fearless at shouting out when I ask a rhetorical question. 00:45:20.810 --> 00:45:22.920 align:middle line:84% Measure numbers, we know, musicians? 00:45:22.920 --> 00:45:23.790 align:middle line:90% Yeah, good, good. 00:45:23.790 --> 00:45:26.090 align:middle line:84% That'll just save time in the class. 00:45:26.090 --> 00:45:29.100 align:middle line:84% So yeah, since musicians know what measure four is, 00:45:29.100 --> 00:45:31.020 align:middle line:90% we'll use measure four that way. 00:45:31.020 --> 00:45:33.840 align:middle line:84% So that's measure four through six. 00:45:33.840 --> 00:45:34.340 align:middle line:90% Great. 00:45:34.340 --> 00:45:36.330 align:middle line:84% So let's start manipulating the score. 00:45:36.330 --> 00:45:37.580 align:middle line:90% Let's say that the-- 00:45:37.580 --> 00:45:40.413 align:middle line:84% AUDIENCE: Four through six, or five through seven? 00:45:40.413 --> 00:45:41.580 align:middle line:90% MICHAEL CUTHBERT: Thank you. 00:45:41.580 --> 00:45:45.990 align:middle line:84% That is whatever a musician thinks measure four is. 00:45:45.990 --> 00:45:48.170 align:middle line:84% So in this particular piece, let's go up 00:45:48.170 --> 00:45:52.080 align:middle line:90% to the beginning of the piece. 00:45:52.080 --> 00:45:54.360 align:middle line:90% Yeah. 00:45:54.360 --> 00:45:57.750 align:middle line:84% What measure is this that the mouse is on? 00:45:57.750 --> 00:45:58.375 align:middle line:90% AUDIENCE: Zero. 00:45:58.375 --> 00:45:59.083 align:middle line:90% AUDIENCE: Pickup. 00:45:59.083 --> 00:46:01.675 align:middle line:84% MICHAEL CUTHBERT: Yeah, Pickup is what a musician would say. 00:46:01.675 --> 00:46:02.800 align:middle line:90% We got to give it a number. 00:46:02.800 --> 00:46:04.510 align:middle line:90% So what number do we like here? 00:46:04.510 --> 00:46:05.010 align:middle line:90% Zero. 00:46:05.010 --> 00:46:07.570 align:middle line:84% So there can be a measure zero if there's a pickup, 00:46:07.570 --> 00:46:10.650 align:middle line:84% but measure one will always be whatever humans, 00:46:10.650 --> 00:46:13.720 align:middle line:84% musicians think measure one should be. 00:46:13.720 --> 00:46:16.720 align:middle line:84% So we can play some other things. 00:46:16.720 --> 00:46:19.540 align:middle line:84% Let's just play around with the score a little bit. 00:46:19.540 --> 00:46:22.660 align:middle line:84% We'll take the soprano, get measure five. 00:46:22.660 --> 00:46:25.540 align:middle line:84% You'll also see some of our display things 00:46:25.540 --> 00:46:28.620 align:middle line:84% don't know how to display the right number. 00:46:28.620 --> 00:46:30.630 align:middle line:84% If you're on MuseScore four, it now 00:46:30.630 --> 00:46:33.540 align:middle line:84% displays the actual correct number. 00:46:33.540 --> 00:46:38.200 align:middle line:84% But so it's measures four through five. 00:46:38.200 --> 00:46:41.080 align:middle line:90% Let's make the fermata red. 00:46:41.080 --> 00:46:42.430 align:middle line:90% So what number? 00:46:42.430 --> 00:46:45.083 align:middle line:84% If I'm getting notes, I'm going to get an index by number. 00:46:45.083 --> 00:46:46.500 align:middle line:84% So we're going to get measure five 00:46:46.500 --> 00:46:48.150 align:middle line:84% and get a particular number on that. 00:46:48.150 --> 00:46:50.220 align:middle line:90% What note number is the fermata? 00:46:50.220 --> 00:46:51.650 align:middle line:90% Shout it out. 00:46:51.650 --> 00:46:52.760 align:middle line:90% AUDIENCE: Three, two? 00:46:52.760 --> 00:46:54.050 align:middle line:90% AUDIENCE: Three. 00:46:54.050 --> 00:46:55.250 align:middle line:90% MICHAEL CUTHBERT: Two. 00:46:55.250 --> 00:46:58.040 align:middle line:84% In this case, musicians don't tend to talk about-- 00:46:58.040 --> 00:46:59.580 align:middle line:84% I guess we do say the second note. 00:46:59.580 --> 00:47:01.200 align:middle line:84% Yeah, yeah, that's really confusing. 00:47:01.200 --> 00:47:03.380 align:middle line:84% So in this case, basically, when we're 00:47:03.380 --> 00:47:07.122 align:middle line:84% using the bracket, square bracket notation, 00:47:07.122 --> 00:47:08.580 align:middle line:84% I'll make this a little bit bigger. 00:47:08.580 --> 00:47:09.455 align:middle line:90% Does that help a bit? 00:47:09.455 --> 00:47:11.720 align:middle line:84% Yeah, we'll think of it-- computer. 00:47:11.720 --> 00:47:18.200 align:middle line:84% So we'll say C. I should call it C-sharp equals, but then you 00:47:18.200 --> 00:47:20.760 align:middle line:84% just get what comments for the rest of the line. 00:47:20.760 --> 00:47:23.120 align:middle line:84% So we'll just call it C, even though it's a C-sharp. 00:47:23.120 --> 00:47:27.030 align:middle line:84% C style-- I think I can just check. 00:47:27.030 --> 00:47:27.560 align:middle line:90% Nope. 00:47:27.560 --> 00:47:31.790 align:middle line:84% That doesn't give me auto-complete-- equals red. 00:47:31.790 --> 00:47:34.600 align:middle line:90% And soprano, show. 00:47:34.600 --> 00:47:42.950 align:middle line:90% 00:47:42.950 --> 00:47:44.250 align:middle line:90% We got the soprano. 00:47:44.250 --> 00:47:47.560 align:middle line:84% I'll scroll back up again so you can see. 00:47:47.560 --> 00:47:48.920 align:middle line:90% We already got the soprano. 00:47:48.920 --> 00:47:52.360 align:middle line:84% We take a particular note, the third note indexed 00:47:52.360 --> 00:47:58.930 align:middle line:84% by 2 from measure five, which is actually 5, and assign it here. 00:47:58.930 --> 00:48:01.210 align:middle line:84% Probably, we could have made different decisions 00:48:01.210 --> 00:48:05.300 align:middle line:84% on how we number things if for Version 2, but whatever. 00:48:05.300 --> 00:48:08.020 align:middle line:84% And then we'll set the style of that note's color to red, 00:48:08.020 --> 00:48:10.720 align:middle line:84% and then show the entire soprano line. 00:48:10.720 --> 00:48:13.660 align:middle line:90% And we get something over here. 00:48:13.660 --> 00:48:15.530 align:middle line:90% I'm half colorblind. 00:48:15.530 --> 00:48:17.300 align:middle line:84% Is there actually a red note up there? 00:48:17.300 --> 00:48:18.710 align:middle line:90% OK, good, good, good. 00:48:18.710 --> 00:48:19.370 align:middle line:90% Where is it? 00:48:19.370 --> 00:48:20.240 align:middle line:90% It's over here. 00:48:20.240 --> 00:48:23.417 align:middle line:84% We'll emphasize that for the camera. 00:48:23.417 --> 00:48:24.250 align:middle line:90% Shine a laser on it. 00:48:24.250 --> 00:48:27.402 align:middle line:90% It becomes more red. 00:48:27.402 --> 00:48:32.290 align:middle line:84% Now, yeah, that's kind of fun play-around things. 00:48:32.290 --> 00:48:34.385 align:middle line:84% Some things we might actually want to do, 00:48:34.385 --> 00:48:35.510 align:middle line:90% let's take the whole piece. 00:48:35.510 --> 00:48:38.002 align:middle line:90% We called it Bach, right? 00:48:38.002 --> 00:48:38.890 align:middle line:90% Zero index,yeah. 00:48:38.890 --> 00:48:40.640 align:middle line:90% Bach's the score object. 00:48:40.640 --> 00:48:42.150 align:middle line:90% Let's analyze its key. 00:48:42.150 --> 00:48:44.760 align:middle line:90% 00:48:44.760 --> 00:48:47.200 align:middle line:90% Analyze is a method. 00:48:47.200 --> 00:48:49.920 align:middle line:84% So we call it like this, parentheses, parentheses, 00:48:49.920 --> 00:48:51.570 align:middle line:90% score key. 00:48:51.570 --> 00:48:56.350 align:middle line:84% So let me say, if you're following along at 80%, 00:48:56.350 --> 00:48:59.130 align:middle line:84% but you just wish I would go a little bit slower, 00:48:59.130 --> 00:49:02.010 align:middle line:84% and then you'd be able to have it, it's going to be fine, 00:49:02.010 --> 00:49:05.500 align:middle line:84% because most of the time, we're going to go slower than this. 00:49:05.500 --> 00:49:08.340 align:middle line:84% This is just the fun beginning of the semester. 00:49:08.340 --> 00:49:11.670 align:middle line:84% If you've never seen a parentheses 00:49:11.670 --> 00:49:13.710 align:middle line:84% or square parentheses, this might 00:49:13.710 --> 00:49:16.740 align:middle line:84% be a class to take next year or next semester 00:49:16.740 --> 00:49:19.800 align:middle line:84% after doing a little bit more programming. 00:49:19.800 --> 00:49:22.960 align:middle line:84% And that's probably more for the camera, people watching at home. 00:49:22.960 --> 00:49:27.210 align:middle line:84% Let's make sure that we just need a little bit of programming 00:49:27.210 --> 00:49:29.472 align:middle line:84% before we're going to go forward. 00:49:29.472 --> 00:49:34.800 align:middle line:84% So the piece as a whole, F-sharp minor. 00:49:34.800 --> 00:49:36.100 align:middle line:90% What do we do now? 00:49:36.100 --> 00:49:37.600 align:middle line:84% I've tried to emphasize it early on, 00:49:37.600 --> 00:49:39.683 align:middle line:84% but we're going to make this a habit in the class. 00:49:39.683 --> 00:49:44.070 align:middle line:84% We always check our bs meter It says 00:49:44.070 --> 00:49:47.090 align:middle line:90% the piece is in F-sharp minor. 00:49:47.090 --> 00:49:50.643 align:middle line:84% Is that at all plausible looking at this piece? 00:49:50.643 --> 00:49:51.810 align:middle line:90% Let's just look at the base. 00:49:51.810 --> 00:49:52.790 align:middle line:90% It's easier. 00:49:52.790 --> 00:49:57.232 align:middle line:84% Well, it looks like an A. It's the three sharps, maybe. 00:49:57.232 --> 00:49:59.120 align:middle line:90% Maybe. 00:49:59.120 --> 00:50:02.480 align:middle line:90% The end, F-sharp. 00:50:02.480 --> 00:50:03.450 align:middle line:90% A-sharp. 00:50:03.450 --> 00:50:05.870 align:middle line:84% Oh, it looks like it ends at F-sharp major. 00:50:05.870 --> 00:50:08.250 align:middle line:84% But then maybe the rest of the piece is in minor, 00:50:08.250 --> 00:50:11.460 align:middle line:84% and then this is just the Bach ending thing. 00:50:11.460 --> 00:50:13.890 align:middle line:84% So to me, it passes the smell test. 00:50:13.890 --> 00:50:15.440 align:middle line:84% It might be other things, but one 00:50:15.440 --> 00:50:19.050 align:middle line:84% of the other things we can do is let's analyze measures. 00:50:19.050 --> 00:50:23.420 align:middle line:84% Oops, we get Bach dot measures, 0. 00:50:23.420 --> 00:50:26.760 align:middle line:84% I haven't spoken out loud while I typed for a while. 00:50:26.760 --> 00:50:27.900 align:middle line:90% 0, comma 2. 00:50:27.900 --> 00:50:30.630 align:middle line:84% So we'll just get the pickup and the first two measures, 00:50:30.630 --> 00:50:32.500 align:middle line:84% and we'll analyze the key of this opening. 00:50:32.500 --> 00:50:36.530 align:middle line:90% 00:50:36.530 --> 00:50:37.640 align:middle line:90% A major. 00:50:37.640 --> 00:50:40.380 align:middle line:84% That feels more right for the opening. 00:50:40.380 --> 00:50:43.460 align:middle line:84% So we know that the piece as a whole is in F-sharp minor, 00:50:43.460 --> 00:50:45.002 align:middle line:84% or at least the computer thinks so, 00:50:45.002 --> 00:50:46.210 align:middle line:90% if the computer got it right. 00:50:46.210 --> 00:50:49.240 align:middle line:90% And it opens an A major. 00:50:49.240 --> 00:50:51.670 align:middle line:84% A logical thing that we might play with next 00:50:51.670 --> 00:50:54.170 align:middle line:90% is, where does it change? 00:50:54.170 --> 00:50:54.890 align:middle line:90% Can we do that? 00:50:54.890 --> 00:50:56.780 align:middle line:84% Can we analyze where key changes are? 00:50:56.780 --> 00:50:59.020 align:middle line:90% We're not going to do that. 00:50:59.020 --> 00:51:00.500 align:middle line:90% Let's get some easy thing. 00:51:00.500 --> 00:51:03.590 align:middle line:84% So we said that not notes I used to get a particular note. 00:51:03.590 --> 00:51:07.850 align:middle line:84% So let's see how many notes there are in this piece. 00:51:07.850 --> 00:51:09.370 align:middle line:90% Len. 00:51:09.370 --> 00:51:11.530 align:middle line:84% Any other language would be dot length, 00:51:11.530 --> 00:51:12.940 align:middle line:90% but we can len Bach dot notes. 00:51:12.940 --> 00:51:14.950 align:middle line:90% And how many notes? 00:51:14.950 --> 00:51:17.830 align:middle line:90% Make a prediction? 00:51:17.830 --> 00:51:18.880 align:middle line:90% Zero. 00:51:18.880 --> 00:51:21.350 align:middle line:84% There are no notes in this piece. 00:51:21.350 --> 00:51:24.820 align:middle line:84% We started with cage 433, and we end with Bach 433. 00:51:24.820 --> 00:51:26.325 align:middle line:90% Why are there no notes? 00:51:26.325 --> 00:51:29.900 align:middle line:90% It's because Bach is a score. 00:51:29.900 --> 00:51:31.870 align:middle line:84% This is all about the representation. 00:51:31.870 --> 00:51:34.270 align:middle line:84% We're currently in box, within box, 00:51:34.270 --> 00:51:36.620 align:middle line:84% within box There are no notes in the score. 00:51:36.620 --> 00:51:37.840 align:middle line:90% There are notes within parts. 00:51:37.840 --> 00:51:39.850 align:middle line:84% No, there are no notes within the parts. 00:51:39.850 --> 00:51:41.410 align:middle line:90% Parts have measures. 00:51:41.410 --> 00:51:42.850 align:middle line:90% There are notes in measures. 00:51:42.850 --> 00:51:47.910 align:middle line:84% So what we'll need to do is, we'll say, take the Bach, 00:51:47.910 --> 00:51:50.560 align:middle line:90% and let's run recurse. 00:51:50.560 --> 00:51:54.190 align:middle line:84% Go into every box like a child at Christmas. 00:51:54.190 --> 00:52:00.450 align:middle line:84% Open every single box and find every single note in there. 00:52:00.450 --> 00:52:02.590 align:middle line:84% So we'll say we can take a score. 00:52:02.590 --> 00:52:06.750 align:middle line:84% We can say open up every box, and to every box, 00:52:06.750 --> 00:52:08.100 align:middle line:90% get me the notes. 00:52:08.100 --> 00:52:09.660 align:middle line:90% And 165. 00:52:09.660 --> 00:52:11.820 align:middle line:90% This passes the smell test. 00:52:11.820 --> 00:52:12.960 align:middle line:90% Let's scroll up. 00:52:12.960 --> 00:52:16.120 align:middle line:84% Don't ask anybody ever to count this, 00:52:16.120 --> 00:52:20.580 align:middle line:84% but that look like about 165 notes to you? 00:52:20.580 --> 00:52:21.720 align:middle line:90% Yeah. 00:52:21.720 --> 00:52:23.662 align:middle line:90% Just order of magnitude. 00:52:23.662 --> 00:52:25.870 align:middle line:84% If it said 1 million notes, would something be wrong? 00:52:25.870 --> 00:52:27.790 align:middle line:84% If it said zero notes, you knew it was something wrong. 00:52:27.790 --> 00:52:30.180 align:middle line:84% So we're always trying to do these kind of smell tests 00:52:30.180 --> 00:52:31.150 align:middle line:90% before we go on. 00:52:31.150 --> 00:52:34.205 align:middle line:84% And the last thing I'm going to do, 00:52:34.205 --> 00:52:35.580 align:middle line:84% because I'm looking at the time-- 00:52:35.580 --> 00:52:40.170 align:middle line:84% and I have only one good habit as a professor. 00:52:40.170 --> 00:52:42.153 align:middle line:84% I really am allergic to keeping you over. 00:52:42.153 --> 00:52:43.820 align:middle line:84% So we're not going to do this, but we'll 00:52:43.820 --> 00:52:48.686 align:middle line:84% do just for fun Bach dot chordify dot show. 00:52:48.686 --> 00:52:55.160 align:middle line:90% 00:52:55.160 --> 00:53:01.490 align:middle line:84% Now we've switched the entire thing into salami slice mode. 00:53:01.490 --> 00:53:05.540 align:middle line:84% So these are some of the things we're going to be doing. 00:53:05.540 --> 00:53:07.260 align:middle line:84% Hopefully, that was good professor mode. 00:53:07.260 --> 00:53:09.390 align:middle line:84% Now here's really evil professor mode. 00:53:09.390 --> 00:53:11.510 align:middle line:84% You're going to forget every tool 00:53:11.510 --> 00:53:13.140 align:middle line:90% that you've just been given. 00:53:13.140 --> 00:53:16.260 align:middle line:84% Analyze key is not available to you. 00:53:16.260 --> 00:53:17.610 align:middle line:90% I'll call it locking it. 00:53:17.610 --> 00:53:20.370 align:middle line:84% It's locked in this little cabinet here. 00:53:20.370 --> 00:53:25.440 align:middle line:84% When you write your own pretty decent key analysis algorithm, 00:53:25.440 --> 00:53:27.300 align:middle line:90% you can unlock this one. 00:53:27.300 --> 00:53:28.820 align:middle line:90% Yours might be better. 00:53:28.820 --> 00:53:31.997 align:middle line:84% I might ask for a license to, hey, can we use that one there. 00:53:31.997 --> 00:53:34.580 align:middle line:84% This is a pretty decent one, but you might write a better one. 00:53:34.580 --> 00:53:38.150 align:middle line:84% When you write your own chordification algorithm, 00:53:38.150 --> 00:53:41.560 align:middle line:84% then chordify will be opened for you. 00:53:41.560 --> 00:53:44.470 align:middle line:84% We'll give you some tools, extracting 00:53:44.470 --> 00:53:46.210 align:middle line:90% things, particular measures. 00:53:46.210 --> 00:53:49.090 align:middle line:84% That was a lot of work, but that kind of sounds boring and boring 00:53:49.090 --> 00:53:51.770 align:middle line:84% in terms of our intellectual development. 00:53:51.770 --> 00:53:56.650 align:middle line:84% But coming up with the tools to be able to chordify a score, 00:53:56.650 --> 00:54:01.120 align:middle line:84% we'll see how many difficult things that come from that. 00:54:01.120 --> 00:54:04.420 align:middle line:84% And the time's about up, so we'll talk 00:54:04.420 --> 00:54:06.290 align:middle line:90% about that in other classes. 00:54:06.290 --> 00:54:08.750 align:middle line:84% As I said, on Friday, I want to get to know you guys. 00:54:08.750 --> 00:54:10.970 align:middle line:84% I want you guys to be able to talk more. 00:54:10.970 --> 00:54:14.260 align:middle line:84% But I wanted today just to blab about what 00:54:14.260 --> 00:54:15.680 align:middle line:90% we might do in this semester. 00:54:15.680 --> 00:54:17.630 align:middle line:90% And I'm happy to have you here. 00:54:17.630 --> 00:54:19.390 align:middle line:90% The numbers look great. 00:54:19.390 --> 00:54:20.870 align:middle line:90% Y'all, come back Friday. 00:54:20.870 --> 00:54:22.610 align:middle line:84% Y'all, make sure that you're registered, 00:54:22.610 --> 00:54:25.800 align:middle line:84% and I'm glad to have everyone in this class. 00:54:25.800 --> 00:54:36.000 align:middle line:90%