mp2i-vms.fr
Compare commits
18 Commits
main
...
refactorin
| Author | SHA1 | Date |
|---|---|---|
 Alexandre
|
b0a6a52057 | |
|
Alexandre
|
ed75cf450d | |
|
Alexandre
|
eb58997fb4 | |
|
Alexandre
|
5d4a9c8c58 | |
|
Alexandre
|
c2cfc429ae | |
|
Alexandre
|
3b6ed078f4 | |
|
Alexandre
|
bb33b03436 | |
|
Alexandre
|
8f035375ab | |
|
Alexandre
|
05b642bdcc | |
|
alexandre
|
64a2a96628 | |
|
Alexandre
|
7615d41c01 | |
|
Alexandre
|
421cddf267 | |
|
Alexandre
|
a59c0c4e08 | |
|
Alexandre
|
02c5579186 | |
|
Alexandre
|
c969501c52 | |
|
Alexandre
|
cba14e4782 | |
|
Alexandre
|
542b6e9996 | |
|
Alexandre
|
9e7c18930f |
|
|
@ -1,2 +1,7 @@
|
|||
*.csv
|
||||
.venv
|
||||
songs/
|
||||
pptx/code/
|
||||
pptx/data/
|
||||
pptx/*.png
|
||||
pptx/*.jpg
|
||||
|
|
@ -1,503 +0,0 @@
|
|||
osu file format v14
|
||||
|
||||
[General]
|
||||
AudioFilename: audio.wav
|
||||
AudioLeadIn: 0
|
||||
PreviewTime: 63278
|
||||
Countdown: 0
|
||||
SampleSet: Normal
|
||||
StackLeniency: 0.5
|
||||
Mode: 0
|
||||
LetterboxInBreaks: 0
|
||||
WidescreenStoryboard: 1
|
||||
|
||||
[Editor]
|
||||
DistanceSpacing: 1.2
|
||||
BeatDivisor: 4
|
||||
GridSize: 32
|
||||
TimelineZoom: 1.8
|
||||
|
||||
[Metadata]
|
||||
Title:Shinzou o Sasageyo! [TV Size]
|
||||
TitleUnicode:心臓を捧げよ! [TV Size]
|
||||
Artist:Linked Horizon
|
||||
ArtistUnicode:Linked Horizon
|
||||
Creator:Monstrata
|
||||
Version:Insane
|
||||
Source:進撃の巨人
|
||||
Tags:revo Attack on Titan shingeki no kyojin season 2 two dedicate all your hearts heart snk aot Haruto haruto_aizawa armin arlert levi erin jaeger mikasa ackerman survey corps opening one theme song sound
|
||||
BeatmapID:1256136
|
||||
BeatmapSetID:593620
|
||||
|
||||
[Difficulty]
|
||||
HPDrainRate:5.5
|
||||
CircleSize:4
|
||||
OverallDifficulty:7.5
|
||||
ApproachRate:9
|
||||
SliderMultiplier:2.2
|
||||
SliderTickRate:1
|
||||
|
||||
[Events]
|
||||
//Background and Video events
|
||||
0,0,"sasageyo.jpg",0,0
|
||||
Video,-150,"sasageyo.flv"
|
||||
//Break Periods
|
||||
//Storyboard Layer 0 (Background)
|
||||
//Storyboard Layer 1 (Fail)
|
||||
//Storyboard Layer 2 (Pass)
|
||||
//Storyboard Layer 3 (Foreground)
|
||||
//Storyboard Sound Samples
|
||||
|
||||
[TimingPoints]
|
||||
560,375,4,2,2,50,1,0
|
||||
560,-200,4,2,2,50,0,0
|
||||
3560,-133.333333333333,4,3,2,65,0,0
|
||||
6560,-166.666666666667,4,3,3,55,0,0
|
||||
17341,-166.666666666667,4,3,3,5,0,0
|
||||
17435,-166.666666666667,4,3,2,55,0,0
|
||||
17716,-166.666666666667,4,3,2,5,0,0
|
||||
17810,-166.666666666667,4,3,2,50,0,0
|
||||
18091,-166.666666666667,4,3,2,5,0,0
|
||||
18185,-166.666666666667,4,3,2,60,0,0
|
||||
18560,-153.846153846154,4,3,2,50,0,0
|
||||
18747,-153.846153846154,4,3,3,60,0,0
|
||||
29060,-133.333333333333,4,3,2,70,0,0
|
||||
29341,-133.333333333333,4,3,2,5,0,0
|
||||
29435,-133.333333333333,4,3,2,70,0,0
|
||||
29716,-133.333333333333,4,3,2,5,0,0
|
||||
29810,-133.333333333333,4,3,2,70,0,0
|
||||
30560,-166.666666666667,4,3,2,55,0,0
|
||||
32810,-166.666666666667,4,3,2,60,0,0
|
||||
33185,-166.666666666667,4,3,2,60,0,0
|
||||
33560,-166.666666666667,4,3,2,55,0,0
|
||||
35810,-153.846153846154,4,3,2,60,0,0
|
||||
36560,-133.333333333333,4,3,2,60,0,0
|
||||
39091,-133.333333333333,4,3,2,5,0,0
|
||||
39185,-133.333333333333,4,3,2,60,0,0
|
||||
39466,-133.333333333333,4,3,2,5,0,0
|
||||
39560,-133.333333333333,4,3,2,60,0,0
|
||||
41060,-100,4,3,2,60,0,0
|
||||
41435,-100,4,3,2,65,0,0
|
||||
41810,-100,4,3,2,70,0,0
|
||||
42185,-97.0873786407767,4,3,2,75,0,0
|
||||
42560,-100,4,3,2,70,0,0
|
||||
45466,-100,4,3,2,5,0,0
|
||||
45560,-166.666666666667,4,3,2,50,0,0
|
||||
48091,-166.666666666667,4,3,2,5,0,0
|
||||
48185,-125,4,3,2,60,0,0
|
||||
48560,-100,4,3,2,70,0,0
|
||||
50060,-125,4,3,2,75,0,0
|
||||
51560,-200,4,3,2,55,0,0
|
||||
54372,-200,4,3,3,55,0,0
|
||||
54560,-200,4,3,2,55,0,0
|
||||
57560,-300,4,3,2,45,0,0
|
||||
60185,-200,4,3,2,55,0,0
|
||||
60560,-133.333333333333,4,3,2,65,0,0
|
||||
62060,-117.647058823529,4,3,2,70,0,0
|
||||
62435,-111.111111111111,4,3,2,70,0,0
|
||||
62810,-105.263157894737,4,3,2,75,0,0
|
||||
63560,-94.1176470588235,4,1,2,70,0,1
|
||||
72560,-100,4,3,2,65,0,0
|
||||
73778,-100,4,3,3,65,0,0
|
||||
73966,-100,4,3,2,65,0,0
|
||||
74060,-133.333333333333,4,3,2,70,0,1
|
||||
74247,-133.333333333333,4,3,3,70,0,1
|
||||
78372,-100,4,3,2,75,0,1
|
||||
78653,-100,4,3,3,75,0,1
|
||||
78747,-100,4,3,2,70,0,1
|
||||
79028,-100,4,3,3,70,0,1
|
||||
79122,-100,4,3,2,70,0,1
|
||||
79403,-100,4,3,3,70,0,1
|
||||
79497,-100,4,3,2,70,0,1
|
||||
80060,-94.1176470588235,4,1,2,75,0,1
|
||||
83528,-94.1176470588235,4,3,3,80,0,1
|
||||
83622,-94.1176470588235,4,1,2,70,0,1
|
||||
86060,-100,4,3,2,80,0,0
|
||||
|
||||
|
||||
[Colours]
|
||||
Combo1 : 222,184,135
|
||||
Combo2 : 255,128,0
|
||||
Combo3 : 0,128,192
|
||||
Combo4 : 255,255,128
|
||||
|
||||
[HitObjects]
|
||||
35,98,560,5,2,0:2:0:0:
|
||||
105,141,747,2,0,L|162:133,1,55,0|0,0:0|0:0,0:0:0:0:
|
||||
124,221,1122,2,0,L|178:213,1,55,2|0,0:0|0:0,0:0:0:0:
|
||||
143,301,1497,2,0,L|197:293,1,55,2|0,0:0|0:0,0:0:0:0:
|
||||
258,237,1872,1,2,0:0:0:0:
|
||||
254,354,2060,5,2,0:1:0:0:
|
||||
314,297,2247,2,0,L|368:304,1,55,2|0,0:0|0:0,0:0:0:0:
|
||||
339,217,2622,2,0,L|393:224,1,55,2|0,0:0|0:0,0:0:0:0:
|
||||
364,137,2997,2,0,L|418:144,1,55,2|0,0:1|0:0,0:0:0:0:
|
||||
389,57,3372,1,2,0:0:0:0:
|
||||
283,156,3560,5,4,0:2:0:0:
|
||||
471,202,3747,2,0,L|478:93,1,82.5000031471253,8|0,0:0|1:0,0:0:0:0:
|
||||
278,73,4122,2,0,L|283:155,1,82.5000031471253,4|0,0:2|0:0,0:0:0:0:
|
||||
212,213,4403,1,8,0:0:0:0:
|
||||
212,213,4497,2,0,L|206:295,1,82.5000031471253,4|4,1:2|2:3,0:0:0:0:
|
||||
277,353,4778,1,8,3:3:0:0:
|
||||
277,353,4872,6,0,L|359:347,1,82.5000031471253,4|0,0:2|2:3,0:0:0:0:
|
||||
410,272,5153,1,8,0:0:0:0:
|
||||
410,272,5247,2,0,P|415:232|404:193,1,82.5000031471254,4|4,0:2|2:3,0:0:0:0:
|
||||
295,242,5528,1,0,2:3:0:0:
|
||||
295,242,5622,2,0,L|359:237,1,41.2500015735627,4|0,0:2|0:0,0:0:0:0:
|
||||
304,163,5810,5,0,1:0:0:0:
|
||||
288,162,5903,1,8,0:0:0:0:
|
||||
273,165,5997,1,4,0:2:0:0:
|
||||
258,171,6091,1,8,0:0:0:0:
|
||||
245,179,6185,1,0,0:0:0:0:
|
||||
234,189,6278,1,0,1:0:0:0:
|
||||
225,202,6372,1,4,0:3:0:0:
|
||||
218,216,6466,1,8,0:0:0:0:
|
||||
215,231,6560,5,4,0:1:0:0:
|
||||
297,307,6747,2,0,L|295:242,1,65.9999979858399,2|2,0:1|0:0,0:0:0:0:
|
||||
218,314,7122,2,0,L|220:379,1,65.9999979858399,2|2,0:2|0:0,0:0:0:0:
|
||||
377,301,7497,2,0,L|374:235,1,65.9999979858399,2|2,0:2|0:0,0:0:0:0:
|
||||
292,176,7872,1,2,0:0:0:0:
|
||||
292,176,7966,1,2,0:0:0:0:
|
||||
292,176,8060,6,0,L|294:241,2,65.9999979858399,10|0|2,0:0|0:0|0:0,0:0:0:0:
|
||||
354,66,8622,2,0,L|356:131,1,65.9999979858399,2|2,0:2|0:0,0:0:0:0:
|
||||
227,131,8997,1,6,0:0:0:0:
|
||||
199,56,9185,1,10,0:0:0:0:
|
||||
199,56,9278,1,10,0:0:0:0:
|
||||
199,56,9372,1,2,0:0:0:0:
|
||||
149,117,9560,5,6,0:0:0:0:
|
||||
175,192,9747,2,0,P|203:208|235:211,1,65.9999979858399,2|2,0:2|0:0,0:0:0:0:
|
||||
359,182,10122,2,0,P|331:165|299:163,1,65.9999979858399,2|2,0:2|0:0,0:0:0:0:
|
||||
309,241,10497,2,0,P|337:257|369:260,1,65.9999979858399,2|2,0:2|0:0,0:0:0:0:
|
||||
493,231,10872,1,2,0:0:0:0:
|
||||
493,231,10966,1,2,0:0:0:0:
|
||||
493,231,11060,6,0,P|464:214|432:211,2,65.9999979858399,6|0|2,0:0|0:0|0:0,0:0:0:0:
|
||||
421,351,11622,2,0,P|438:323|441:291,1,65.9999979858399,2|2,0:2|0:0,0:0:0:0:
|
||||
352,219,11997,1,10,0:0:0:0:
|
||||
287,265,12185,1,4,0:2:0:0:
|
||||
364,302,12372,1,2,0:0:0:0:
|
||||
364,302,12466,1,2,0:0:0:0:
|
||||
364,302,12560,5,4,0:2:0:0:
|
||||
344,140,12747,2,0,L|350:59,1,65.9999979858399,0|2,0:0|0:0,0:0:0:0:
|
||||
266,156,13122,2,0,L|271:91,1,65.9999979858399,2|10,0:2|0:0,0:0:0:0:
|
||||
188,172,13497,2,0,L|192:106,1,65.9999979858399,2|2,0:2|0:0,0:0:0:0:
|
||||
109,187,13872,1,2,0:0:0:0:
|
||||
109,187,13966,1,2,0:0:0:0:
|
||||
109,187,14060,6,0,L|113:121,2,65.9999979858399,6|2|2,0:0|0:0|0:0,0:0:0:0:
|
||||
101,265,14622,2,0,P|69:255|46:233,1,65.9999979858399,2|2,0:2|0:1,0:0:0:0:
|
||||
31,154,14997,1,10,0:0:0:0:
|
||||
112,121,15185,2,0,L|108:187,1,65.9999979858399,2|2,0:2|0:1,0:0:0:0:
|
||||
109,187,15466,1,2,0:0:0:0:
|
||||
109,187,15560,5,10,0:0:0:0:
|
||||
191,93,15747,2,0,L|187:159,1,65.9999979858399,2|2,0:2|0:0,0:0:0:0:
|
||||
253,41,16122,1,10,0:0:0:0:
|
||||
253,41,16216,1,10,0:0:0:0:
|
||||
252,40,16310,2,0,P|267:70|268:103,1,65.9999979858399,2|0,0:1|0:0,0:0:0:0:
|
||||
342,130,16685,2,0,P|327:159|326:192,1,65.9999979858399,2|2,3:0|0:0,0:0:0:0:
|
||||
376,254,16966,1,2,0:0:0:0:
|
||||
376,254,17060,6,0,P|430:269|490:240,1,98.9999969787599,4|0,0:2|0:0,0:0:0:0:
|
||||
472,163,17435,2,0,P|457:217|486:277,1,98.9999969787599,8|0,0:2|0:0,0:0:0:0:
|
||||
504,360,17810,2,0,P|450:345|390:374,1,98.9999969787599,10|0,1:2|0:2,0:0:0:0:
|
||||
311,384,18185,2,0,P|326:330|297:270,1,98.9999969787599,12|8,0:2|0:0,0:0:0:0:
|
||||
256,192,18560,5,4,0:1:0:0:
|
||||
175,164,18747,2,0,P|172:196|203:261,1,71.4999967269899,2|2,0:1|0:0,0:0:0:0:
|
||||
337,220,19122,2,0,P|340:188|309:123,1,71.4999967269899,2|2,0:2|0:0,0:0:0:0:
|
||||
256,19,19497,1,2,0:2:0:0:
|
||||
255,106,19685,1,2,0:0:0:0:
|
||||
331,63,19872,1,2,0:2:0:0:
|
||||
180,62,20060,5,2,0:2:0:0:
|
||||
256,192,20247,2,0,L|350:201,1,71.4999967269899,0|2,0:0|0:0,0:0:0:0:
|
||||
402,69,20622,2,0,L|330:62,1,71.4999967269899,2|2,0:2|0:0,0:0:0:0:
|
||||
255,106,20997,1,0,0:0:0:0:
|
||||
405,218,21185,2,0,L|411:146,1,71.4999967269899,12|2,0:2|0:0,0:0:0:0:
|
||||
256,192,21560,5,12,0:2:0:0:
|
||||
228,273,21747,2,0,P|260:276|325:245,1,71.4999967269899,2|2,0:2|0:0,0:0:0:0:
|
||||
284,111,22122,2,0,P|252:108|187:139,1,71.4999967269899,2|2,0:2|0:0,0:0:0:0:
|
||||
83,192,22497,1,0,0:0:0:0:
|
||||
170,193,22685,1,2,0:2:0:0:
|
||||
127,117,22872,1,0,0:0:0:0:
|
||||
126,268,23060,5,2,0:0:0:0:
|
||||
256,192,23247,2,0,P|294:186|349:216,1,71.4999967269899,2|2,0:2|0:0,0:0:0:0:
|
||||
308,355,23622,2,0,P|272:358|239:346,1,71.4999967269899,2|2,0:2|0:0,0:0:0:0:
|
||||
284,273,23997,1,2,0:0:0:0:
|
||||
392,373,24185,1,4,0:2:0:0:
|
||||
409,375,24278,1,0,1:0:0:0:
|
||||
427,375,24372,1,0,1:0:0:0:
|
||||
444,372,24466,1,0,1:0:0:0:
|
||||
461,365,24560,5,12,0:2:0:0:
|
||||
425,219,24747,2,0,L|418:291,1,71.4999967269899,2|2,0:2|0:0,0:0:0:0:
|
||||
284,273,25122,2,0,L|350:245,1,71.4999967269899,2|2,0:1|0:0,0:0:0:0:
|
||||
401,368,25497,2,0,L|343:325,1,71.4999967269899,2|2,0:2|0:0,0:0:0:0:
|
||||
229,335,25872,1,0,1:0:0:0:
|
||||
229,335,25966,1,0,1:0:0:0:
|
||||
229,335,26060,6,0,L|286:377,1,71.4999967269899,12|8,0:2|0:0,0:0:0:0:
|
||||
284,273,26435,1,2,0:0:0:0:
|
||||
151,303,26622,2,0,P|145:339|154:373,1,71.4999967269899,2|2,0:1|0:0,0:0:0:0:
|
||||
208,239,26997,2,0,P|212:199|197:156,1,71.4999967269899,2|2,0:2|0:0,0:0:0:0:
|
||||
57,217,27372,1,2,0:0:0:0:
|
||||
57,217,27466,1,2,0:0:0:0:
|
||||
57,217,27560,6,0,L|128:209,1,71.4999967269899,4|8,0:2|0:0,0:0:0:0:
|
||||
209,82,27935,1,2,0:0:0:0:
|
||||
132,122,28122,1,2,0:1:0:0:
|
||||
281,127,28310,2,0,P|319:120|345:100,1,71.4999967269899,10|0,0:2|0:0,0:0:0:0:
|
||||
284,42,28685,1,2,0:0:0:0:
|
||||
132,122,28872,1,8,0:0:0:0:
|
||||
270,247,29060,6,0,P|288:187|281:126,1,123.750004720688,4|0,0:2|0:0,0:0:0:0:
|
||||
183,57,29435,2,0,P|224:102|280:126,1,123.750004720688,4|0,1:3|0:0,0:0:0:0:
|
||||
342,7,29810,6,0,P|336:47|347:86,1,82.5000031471254,10|2,0:2|0:1,0:0:0:0:
|
||||
441,118,30091,1,8,0:2:0:0:
|
||||
441,118,30185,2,0,P|447:77|436:38,1,82.5000031471254,4|0,0:2|2:0,0:0:0:0:
|
||||
366,182,30560,5,4,0:1:0:0:
|
||||
444,207,30747,1,0,0:0:0:0:
|
||||
408,340,30935,1,8,0:0:0:0:
|
||||
383,262,31122,1,0,0:0:0:0:
|
||||
250,299,31310,1,0,0:0:0:0:
|
||||
328,324,31497,1,2,0:0:0:0:
|
||||
289,180,31685,2,0,L|310:243,1,65.9999979858399,8|0,0:0|0:0,0:0:0:0:
|
||||
388,120,32060,6,0,L|367:183,1,65.9999979858399,2|0,0:2|0:0,0:0:0:0:
|
||||
278,103,32435,1,8,0:0:0:0:
|
||||
225,227,32622,1,2,0:0:0:0:
|
||||
149,142,32810,2,0,L|215:148,1,65.9999979858399,4|0,0:2|0:0,0:0:0:0:
|
||||
142,221,33185,2,0,L|76:215,1,65.9999979858399,12|0,0:2|0:0,0:0:0:0:
|
||||
132,65,33560,5,4,0:2:0:0:
|
||||
149,142,33747,1,0,0:0:0:0:
|
||||
209,48,33935,1,8,0:0:0:0:
|
||||
226,125,34123,1,2,0:0:0:0:
|
||||
125,24,34310,5,2,0:2:0:0:
|
||||
108,148,34497,1,2,0:0:0:0:
|
||||
249,41,34684,1,8,0:0:0:0:
|
||||
232,165,34872,1,2,0:0:0:0:
|
||||
173,55,35060,5,4,0:3:0:0:
|
||||
286,110,35247,1,2,0:0:0:0:
|
||||
163,132,35435,1,8,0:0:0:0:
|
||||
249,41,35622,1,2,0:0:0:0:
|
||||
232,165,35810,2,0,P|211:193|176:209,1,71.4999967269899,4|0,0:2|0:0,0:0:0:0:
|
||||
151,336,36185,1,12,0:0:0:0:
|
||||
159,320,36278,1,2,0:1:0:0:
|
||||
172,307,36372,1,2,0:2:0:0:
|
||||
187,298,36466,1,2,0:2:0:0:
|
||||
204,293,36560,5,4,0:2:0:0:
|
||||
230,379,36747,1,2,0:0:0:0:
|
||||
267,229,36935,1,8,0:0:0:0:
|
||||
292,314,37122,1,2,0:0:0:0:
|
||||
179,208,37310,1,2,0:2:0:0:
|
||||
117,273,37497,1,2,0:0:0:0:
|
||||
154,123,37685,1,8,0:0:0:0:
|
||||
92,187,37872,1,2,0:0:0:0:
|
||||
242,144,38060,5,4,0:3:0:0:
|
||||
67,102,38247,1,2,0:0:0:0:
|
||||
267,229,38435,1,8,0:0:0:0:
|
||||
217,58,38622,1,2,0:0:0:0:
|
||||
217,58,38716,1,2,0:0:0:0:
|
||||
216,58,38810,6,0,P|273:36|333:50,1,123.750004720688,4|0,0:2|0:0,0:0:0:0:
|
||||
341,167,39185,2,0,P|354:107|333:50,1,123.750004720688,12|0,0:2|0:0,0:0:0:0:
|
||||
242,144,39560,5,4,0:2:0:0:
|
||||
267,229,39747,1,2,0:0:0:0:
|
||||
118,272,39935,1,8,0:0:0:0:
|
||||
180,208,40122,1,2,0:0:0:0:
|
||||
292,314,40310,1,2,0:2:0:0:
|
||||
267,229,40497,1,2,0:0:0:0:
|
||||
179,378,40685,1,8,0:0:0:0:
|
||||
204,293,40872,1,2,0:1:0:0:
|
||||
380,335,41060,6,0,P|426:298|396:263,1,110,12|0,1:2|0:0,0:0:0:0:
|
||||
393,233,41341,1,4,0:0:0:0:
|
||||
393,233,41435,2,0,P|439:196|409:161,1,110,8|0,1:2|0:0,0:0:0:0:
|
||||
406,131,41716,1,0,0:0:0:0:
|
||||
406,131,41810,2,0,P|426:97|414:17,1,110,12|2,1:2|0:1,0:0:0:0:
|
||||
347,70,42091,1,2,0:1:0:0:
|
||||
347,70,42185,2,0,L|399:251,1,169.949998573723,12|0,0:0|0:0,0:0:0:0:
|
||||
335,298,42560,5,12,0:2:0:0:
|
||||
184,118,42747,1,2,0:1:0:0:
|
||||
193,353,42935,1,10,0:2:0:0:
|
||||
271,131,43122,1,2,0:1:0:0:
|
||||
39,174,43310,1,8,0:0:0:0:
|
||||
248,284,43497,1,0,0:0:0:0:
|
||||
248,284,43591,1,0,0:0:0:0:
|
||||
248,284,43685,1,12,0:2:0:0:
|
||||
238,49,43872,1,0,0:0:0:0:
|
||||
238,49,43965,1,4,0:3:0:0:
|
||||
238,49,44059,1,4,0:2:0:0:
|
||||
161,270,44247,5,2,0:1:0:0:
|
||||
325,62,44434,1,10,0:0:0:0:
|
||||
126,187,44622,1,4,0:0:0:0:
|
||||
335,298,44810,1,8,0:0:0:0:
|
||||
335,298,44903,2,0,P|387:290|426:237,1,110,8|2,0:0|0:0,0:0:0:0:
|
||||
346,189,45185,2,0,L|335:299,1,110,12|0,0:2|0:0,0:0:0:0:
|
||||
263,226,45560,6,0,P|232:216|200:216,1,65.9999979858399,4|0,0:1|0:0,0:0:0:0:
|
||||
117,299,45935,2,0,P|127:267|127:235,1,65.9999979858399,2|0,0:0|0:0,0:0:0:0:
|
||||
44,152,46310,2,0,P|75:162|107:162,1,65.9999979858399,8|0,0:0|0:0,0:0:0:0:
|
||||
190,79,46685,2,0,P|179:110|179:142,1,65.9999979858399,2|0,0:1|0:0,0:0:0:0:
|
||||
297,145,47060,6,0,P|239:119|179:142,1,131.99999597168,4|0,0:0|0:0,0:0:0:0:
|
||||
237,206,47622,1,2,0:0:0:0:
|
||||
357,206,47810,2,0,L|387:136,1,65.9999979858399,12|0,0:2|0:0,0:0:0:0:
|
||||
289,58,48185,6,0,L|298:150,1,88,2|0,0:1|0:0,0:0:0:0:
|
||||
435,71,48560,6,0,P|365:32|289:57,1,165,12|0,0:2|0:0,0:0:0:0:
|
||||
151,161,48935,2,0,P|227:185|297:147,1,165,4|0,0:0|1:0,0:0:0:0:
|
||||
165,75,49310,1,10,0:0:0:0:
|
||||
120,327,49497,1,0,0:0:0:0:
|
||||
360,239,49685,1,8,0:0:0:0:
|
||||
99,214,49872,1,0,0:0:0:0:
|
||||
251,0,50060,6,0,L|246:103,1,88,12|0,1:2|0:0,0:0:0:0:
|
||||
329,196,50341,1,0,0:0:0:0:
|
||||
329,196,50435,2,0,L|333:108,1,88,8|0,1:3|0:0,0:0:0:0:
|
||||
417,161,50716,1,0,0:0:0:0:
|
||||
417,161,50810,6,0,L|488:150,1,44,4|0,1:2|1:0,0:0:0:0:
|
||||
451,47,50997,2,0,L|407:53,1,44,8|0,1:3|1:0,0:0:0:0:
|
||||
412,171,51185,2,0,L|483:160,1,44,2|0,1:0|1:0,0:0:0:0:
|
||||
446,57,51372,2,0,L|402:63,1,44,8|0,1:3|1:3,0:0:0:0:
|
||||
343,220,51560,6,0,P|395:255|452:246,1,110,4|0,0:1|0:0,0:0:0:0:
|
||||
412,171,52122,1,2,0:0:0:0:
|
||||
267,185,52310,2,0,P|262:247|298:292,1,110,8|2,0:0|0:1,0:0:0:0:
|
||||
343,220,52872,1,2,0:0:0:0:
|
||||
488,136,53060,6,0,P|492:198|456:243,1,110,2|0,0:2|0:0,0:0:0:0:
|
||||
412,171,53622,1,2,0:0:0:0:
|
||||
259,79,53810,2,0,P|254:141|290:186,1,110,12|2,0:0|0:1,0:0:0:0:
|
||||
343,122,54372,1,2,0:0:0:0:
|
||||
343,122,54466,1,8,0:0:0:0:
|
||||
343,122,54560,6,0,P|390:109|458:144,1,110,4|0,0:2|0:0,0:0:0:0:
|
||||
389,194,55122,1,2,0:0:0:0:
|
||||
169,122,55310,2,0,P|122:109|54:144,1,110,8|2,0:0|0:1,0:0:0:0:
|
||||
123,194,55872,1,2,0:0:0:0:
|
||||
256,152,56060,6,0,L|256:32,1,110,0|2,0:0|0:1,0:0:0:0:
|
||||
180,76,56622,1,2,0:0:0:0:
|
||||
332,76,56810,2,0,L|332:146,1,55,12|0,0:0|0:0,0:0:0:0:
|
||||
256,207,57185,2,0,L|256:152,1,55,12|2,0:2|0:2,0:0:0:0:
|
||||
309,271,57560,6,0,L|399:271,1,73.3333333333333,4|0,0:1|0:0,0:0:0:0:
|
||||
423,206,58122,1,2,0:2:0:0:
|
||||
423,206,58310,2,0,P|447:207|474:201,1,36.6666666666667,0|0,0:0|0:0,0:0:0:0:
|
||||
440,130,58685,2,0,P|441:105|435:79,1,36.6666666666667,0|2,0:0|0:2,0:0:0:0:
|
||||
364,113,59060,6,0,P|337:111|283:136,1,73.3333333333333,2|0,0:0|0:0,0:0:0:0:
|
||||
238,179,59622,1,0,0:0:0:0:
|
||||
238,179,59810,2,0,P|251:204|252:224,1,36.6666666666667,2|0,0:2|0:0,0:0:0:0:
|
||||
123,229,60185,6,0,L|178:224,1,55,8|2,0:0|2:0,0:0:0:0:
|
||||
227,286,60466,1,2,0:1:0:0:
|
||||
227,286,60560,6,0,L|313:294,1,82.5000031471254,12|0,0:2|0:0,0:0:0:0:
|
||||
384,260,60841,1,0,0:0:0:0:
|
||||
384,260,60935,2,0,L|376:342,1,82.5000031471254,2|0,0:1|1:0,0:0:0:0:
|
||||
454,365,61216,1,2,0:1:0:0:
|
||||
454,365,61310,2,0,L|460:292,1,41.2500015735627,12|0,0:2|0:0,0:0:0:0:
|
||||
505,237,61497,2,0,L|464:241,1,41.2500015735627,0|0,1:0|0:0,0:0:0:0:
|
||||
366,175,61685,2,0,L|408:179,1,41.2500015735627,8|0,1:3|0:0,0:0:0:0:
|
||||
438,61,61872,2,0,L|442:103,1,41.2500015735627,2|0,0:1|0:0,0:0:0:0:
|
||||
357,268,62060,6,0,L|366:175,1,93.500001783371,12|0,0:2|0:0,0:0:0:0:
|
||||
220,120,62341,1,0,0:0:0:0:
|
||||
220,120,62435,6,0,L|206:258,1,98.9999969787599,12|0,0:2|0:0,0:0:0:0:
|
||||
284,241,62716,1,0,0:0:0:0:
|
||||
284,241,62810,6,0,L|296:116,1,104.499997209549,12|0,0:2|0:0,0:0:0:0:
|
||||
147,93,63091,1,0,0:0:0:0:
|
||||
147,93,63185,6,0,L|134:224,1,104.499997209549,12|2,0:2|0:1,0:0:0:0:
|
||||
210,219,63466,1,8,0:0:0:0:
|
||||
210,219,63560,6,0,L|225:67,1,116.875002229214,4|2,0:1|0:1,0:0:0:0:
|
||||
294,137,63841,1,0,0:0:0:0:
|
||||
294,137,63935,2,0,L|371:137,1,58.4375011146069,2|0,0:2|0:0,0:0:0:0:
|
||||
328,209,64122,2,0,L|378:238,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
325,294,64310,6,0,P|303:363|259:384,1,116.875002229214,2|10,0:0|0:3,0:0:0:0:
|
||||
191,305,64591,1,0,0:0:0:0:
|
||||
191,305,64685,2,0,L|250:308,1,58.4375011146069,0|0,0:0|3:0,0:0:0:0:
|
||||
258,232,64872,2,0,L|261:173,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
185,200,65060,6,0,P|169:270|197:310,1,116.875002229214,12|8,0:2|0:3,0:0:0:0:
|
||||
141,362,65341,1,0,0:0:0:0:
|
||||
141,362,65435,2,0,L|82:364,1,58.4375011146069,0|0,0:0|3:0,0:0:0:0:
|
||||
41,292,65622,2,0,L|100:290,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
141,143,65810,2,0,L|82:140,1,58.4375011146069,10|0,0:2|0:0,0:0:0:0:
|
||||
42,213,65997,2,0,L|100:215,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
265,93,66185,5,4,0:2:0:0:
|
||||
227,70,66278,1,0,3:0:0:0:
|
||||
184,74,66372,1,8,0:3:0:0:
|
||||
151,102,66466,1,0,0:0:0:0:
|
||||
141,143,66560,6,0,L|290:169,1,116.875002229214,4|2,0:2|0:0,0:0:0:0:
|
||||
294,228,66841,1,0,0:0:0:0:
|
||||
294,228,66935,2,0,L|400:209,1,58.4375011146069,0|0,3:0|3:0,0:0:0:0:
|
||||
386,134,67122,2,0,L|329:145,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
369,290,67310,6,0,P|416:258|424:202,1,116.875002229214,12|2,0:2|0:0,0:0:0:0:
|
||||
464,137,67591,1,0,0:0:0:0:
|
||||
464,137,67685,2,0,L|521:147,1,58.4375011146069,0|0,3:0|3:0,0:0:0:0:
|
||||
486,57,67872,2,0,L|429:68,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
271,114,68060,6,0,L|423:140,1,116.875002229214,12|8,0:2|0:3,0:0:0:0:
|
||||
387,5,68341,1,0,0:0:0:0:
|
||||
387,5,68435,2,0,P|366:26|356:53,1,58.4375011146069,0|0,3:0|3:0,0:0:0:0:
|
||||
427,196,68622,2,0,P|447:175|458:148,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
271,114,68810,6,0,L|297:266,1,116.875002229214,12|0,0:2|0:0,0:0:0:0:
|
||||
354,303,69091,1,0,0:0:0:0:
|
||||
354,303,69185,2,0,L|365:246,1,58.4375011146069,12|0,0:2|0:0,0:0:0:0:
|
||||
427,196,69372,2,0,L|456:246,1,58.4375011146069,8|8,0:3|0:3,0:0:0:0:
|
||||
405,369,69560,6,0,P|350:387|296:363,1,116.875002229214,12|2,0:2|0:0,0:0:0:0:
|
||||
202,297,69841,1,0,0:0:0:0:
|
||||
202,297,69935,2,0,P|228:285|257:285,1,58.4375011146069,0|0,3:0|3:0,0:0:0:0:
|
||||
214,374,70122,2,0,P|184:374|158:362,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
142,210,70310,6,0,P|189:242|202:298,1,116.875002229214,12|10,0:2|0:3,0:0:0:0:
|
||||
84,331,70591,1,0,0:0:0:0:
|
||||
84,331,70685,2,0,P|100:306|124:289,1,58.4375011146069,0|0,3:0|3:0,0:0:0:0:
|
||||
14,219,70872,2,0,P|43:220|70:233,1,58.4375011146069,8|2,0:3|0:3,0:0:0:0:
|
||||
154,94,71060,6,0,L|137:255,1,116.875002229214,12|10,0:2|0:3,0:0:0:0:
|
||||
289,216,71341,1,0,0:0:0:0:
|
||||
289,216,71435,2,0,L|295:157,1,58.4375011146069,0|0,3:0|3:0,0:0:0:0:
|
||||
222,138,71622,2,0,L|216:197,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
340,277,71810,6,0,L|392:251,1,58.4375011146069,12|0,0:2|0:0,0:0:0:0:
|
||||
347,132,71997,2,0,L|278:166,1,58.4375011146069,2|0,0:0|3:0,0:0:0:0:
|
||||
211,323,72185,1,8,0:3:0:0:
|
||||
247,345,72278,1,10,0:3:0:0:
|
||||
291,343,72372,1,12,0:3:0:0:
|
||||
325,318,72466,1,8,0:2:0:0:
|
||||
340,277,72560,5,14,0:2:0:0:
|
||||
340,277,73310,5,2,0:2:0:0:
|
||||
283,212,73497,1,2,3:0:0:0:
|
||||
294,187,73590,1,2,3:1:0:0:
|
||||
290,161,73684,1,6,0:0:0:0:
|
||||
270,142,73778,1,8,1:3:0:0:
|
||||
243,138,73872,1,8,1:3:0:0:
|
||||
220,151,73965,1,2,1:0:0:0:
|
||||
208,175,74059,5,4,0:1:0:0:
|
||||
340,277,74247,2,0,L|440:261,1,82.5000031471254,2|2,0:2|0:0,0:0:0:0:
|
||||
424,98,74622,2,0,L|342:111,1,82.5000031471254,2|2,0:2|0:0,0:0:0:0:
|
||||
341,194,74997,2,0,L|423:181,1,82.5000031471254,2|2,0:2|0:0,0:0:0:0:
|
||||
425,15,75372,1,8,0:0:0:0:
|
||||
273,149,75560,5,4,0:2:0:0:
|
||||
495,225,75747,2,0,P|505:180|491:131,1,82.5000031471254,2|2,0:2|0:0,0:0:0:0:
|
||||
276,69,76122,2,0,P|266:114|280:163,1,82.5000031471254,10|2,0:2|0:0,0:0:0:0:
|
||||
349,111,76497,2,0,P|359:66|345:17,1,82.5000031471254,8|2,0:2|0:0,0:0:0:0:
|
||||
201,25,76872,1,10,0:2:0:0:
|
||||
248,264,77060,5,2,0:0:0:0:
|
||||
430,101,77247,2,0,L|338:112,1,82.5000031471254,10|2,0:2|0:0,0:0:0:0:
|
||||
166,274,77622,2,0,L|247:264,1,82.5000031471254,10|2,0:2|0:0,0:0:0:0:
|
||||
212,55,77997,2,0,L|223:147,1,82.5000031471254,10|0,0:2|1:0,0:0:0:0:
|
||||
387,345,78372,6,0,L|374:237,1,110,14|0,0:2|0:0,0:0:0:0:
|
||||
469,285,78653,1,2,0:0:0:0:
|
||||
469,285,78747,2,0,L|481:176,1,110,12|0,0:2|0:0,0:0:0:0:
|
||||
397,130,79028,1,2,0:0:0:0:
|
||||
397,130,79122,2,0,L|497:87,1,110,12|0,0:2|0:0,0:0:0:0:
|
||||
311,83,79403,1,2,0:0:0:0:
|
||||
311,83,79497,2,0,P|299:136|318:187,1,110,12|0,0:2|1:0,0:0:0:0:
|
||||
378,280,79778,1,8,0:3:0:0:
|
||||
378,280,79872,2,0,L|283:272,1,55,12|8,0:2|1:2,0:0:0:0:
|
||||
287,346,80060,6,0,L|192:354,1,58.4375011146069,6|0,0:1|0:0,0:0:0:0:
|
||||
152,347,80247,2,0,P|206:318|225:257,1,116.875002229214,10|0,0:2|0:0,0:0:0:0:
|
||||
219,188,80528,1,0,0:0:0:0:
|
||||
219,188,80622,6,0,P|165:217|146:278,1,116.875002229214,12|0,0:2|0:0,0:0:0:0:
|
||||
79,303,80903,1,0,3:0:0:0:
|
||||
79,303,80997,2,0,L|20:296,1,58.4375011146069,10|0,0:2|0:0,0:0:0:0:
|
||||
78,226,81185,2,0,L|22:184,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
115,158,81372,2,0,L|87:94,1,58.4375011146069,10|0,0:2|0:0,0:0:0:0:
|
||||
187,59,81560,6,0,L|181:118,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
219,188,81747,2,0,B|274:216|274:170|336:200,1,116.875002229214,10|2,0:2|0:3,0:0:0:0:
|
||||
336,305,82028,1,0,0:0:0:0:
|
||||
336,305,82122,2,0,P|314:252|332:197,1,116.875002229214,12|0,1:2|0:0,0:0:0:0:
|
||||
389,249,82403,1,0,3:0:0:0:
|
||||
389,249,82497,2,0,L|470:253,1,58.4375011146069,10|0,0:2|0:0,0:0:0:0:
|
||||
399,174,82685,2,0,L|471:137,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
371,105,82872,2,0,L|415:37,1,58.4375011146069,10|0,0:2|0:0,0:0:0:0:
|
||||
315,0,83060,6,0,L|313:59,1,58.4375011146069,2|0,0:0|0:0,0:0:0:0:
|
||||
208,55,83247,2,0,P|262:34|313:58,1,116.875002229214,10|2,0:2|0:3,0:0:0:0:
|
||||
309,167,83528,1,8,1:3:0:0:
|
||||
309,167,83622,2,0,P|255:187|204:164,1,116.875002229214,12|0,0:2|0:0,0:0:0:0:
|
||||
132,135,83903,1,0,3:0:0:0:
|
||||
132,135,83997,2,0,L|110:194,1,58.4375011146069,10|0,0:2|0:0,0:0:0:0:
|
||||
35,197,84185,2,0,L|56:256,1,58.4375011146069,8|0,0:3|0:0,0:0:0:0:
|
||||
16,318,84372,6,0,P|72:326|119:294,1,116.875002229214,14|0,0:2|0:0,0:0:0:0:
|
||||
309,133,84747,2,0,P|255:112|202:134,1,116.875002229214,10|0,0:2|0:0,0:0:0:0:
|
||||
392,294,85122,2,0,P|439:326|495:318,1,116.875002229214,14|0,0:2|0:0,0:0:0:0:
|
||||
257,187,85497,1,10,0:3:0:0:
|
||||
249,205,85590,1,0,0:0:0:0:
|
||||
249,224,85684,1,0,0:0:0:0:
|
||||
255,242,85778,1,0,0:0:0:0:
|
||||
262,260,85872,1,8,0:3:0:0:
|
||||
263,280,85965,1,0,0:0:0:0:
|
||||
256,298,86059,5,12,1:2:0:0:
|
||||
|
|
@ -0,0 +1,472 @@
|
|||
from math import *
|
||||
import numpy as np
|
||||
import scipy as scp
|
||||
from scipy.io import wavfile
|
||||
import matplotlib.pyplot as plt
|
||||
import subprocess
|
||||
import heapq
|
||||
from pathlib import Path
|
||||
from time import sleep
|
||||
import datetime
|
||||
|
||||
def is_data_stereo(raw_global_data:list) -> bool:
|
||||
"""
|
||||
self-explainatory
|
||||
"""
|
||||
try:
|
||||
assert(raw_global_data[0][0])
|
||||
except IndexError:
|
||||
return False
|
||||
except AssertionError:
|
||||
return True
|
||||
return True
|
||||
|
||||
def dist_to_integer(x):
|
||||
ent = np.floor(x)
|
||||
if(ent < 0.5):
|
||||
return ent
|
||||
else:
|
||||
return (1-ent)
|
||||
|
||||
def is_note_within(fr1, fr2):
|
||||
if(fr1 > fr2):
|
||||
return (fr1/fr2 <= NOTE_DIST or dist_to_integer(fr1/fr2) >= OCTAVE_DIST) # same tone or octave
|
||||
else:
|
||||
return (fr2/fr1 <= NOTE_DIST or dist_to_integer(fr2/fr1) >= OCTAVE_DIST)
|
||||
|
||||
def keep_highest(song_name, offset, songlen, segsize, count, output_name, minfreq=110, maxfreq=5000, ampthr=250, canPlot=True, writeO = True):
|
||||
'''
|
||||
INPUT : data relative to music + config about the analysis
|
||||
OUTPUT :
|
||||
* a list of timings : it contains floats (representing circles) and couple of floats (representing sliders) (e.g. [float, float])
|
||||
* a list of amplitudes relative to timings
|
||||
'''
|
||||
# extracting data from cropped song
|
||||
sample_rate, raw_song_data = wavfile.read(song_name)
|
||||
blit = int(sample_rate*segsize) # Te
|
||||
|
||||
song_data = [0 for i in range(len(raw_song_data))]
|
||||
|
||||
id_start = int(offset*sample_rate)
|
||||
id_end = min(len(raw_song_data), int((offset+songlen)*sample_rate))
|
||||
|
||||
a = 0
|
||||
if(is_data_stereo(raw_song_data)):
|
||||
print("Converting to mono...")
|
||||
for x in range(id_start, id_end):
|
||||
song_data[x] = raw_song_data[x][0]/2 + raw_song_data[x][1]/2
|
||||
|
||||
if(x % (int(len(raw_song_data)/100)) == 0):
|
||||
print(a, "/ 100")
|
||||
a += 1
|
||||
else:
|
||||
song_data = raw_song_data
|
||||
|
||||
print("\nSampleRate : ", sample_rate)
|
||||
print("SegSize : ", blit)
|
||||
|
||||
# calculate the frequencies associated to the FFTs
|
||||
pfreq = scp.fft.rfftfreq(blit, 1/sample_rate)
|
||||
|
||||
# left boundary of segment to crop
|
||||
current_time = offset
|
||||
|
||||
# list of FFTs
|
||||
fft_list = []
|
||||
fft_list_untouched = []
|
||||
|
||||
# number of samples
|
||||
k = 0
|
||||
|
||||
print("Retrieving freqs from", offset, "to", songlen+offset, "...")
|
||||
while(current_time < songlen+offset-segsize):
|
||||
# index corresponding to left boundary
|
||||
left_id = int(current_time*sample_rate)
|
||||
|
||||
# index corresponding to right boundary
|
||||
right_id = int((current_time+segsize)*sample_rate)
|
||||
|
||||
# calculate the fft, append it to fft_list
|
||||
pff = scp.fft.rfft(song_data[int(current_time*sample_rate):int(sample_rate*(current_time+segsize))])
|
||||
fft_list.append(pff)
|
||||
fft_list_untouched.append([ee for ee in pff])
|
||||
|
||||
# just to avoid what causes 0.1 + 0.1 == 0.2 to be False
|
||||
k += 1
|
||||
current_time = offset + k*segsize
|
||||
#print(current_time)
|
||||
|
||||
print("\n\nSegSize :", segsize, "\nFFT :", len(fft_list), "\nFFT[0] :", len(fft_list[0]), "\npfreq :", len(pfreq), "\n\n")
|
||||
|
||||
# -------------------------------------------- Clean song -------------------------------------------- #
|
||||
pfreq_minid = 0
|
||||
pfreq_maxid = len(pfreq) -1
|
||||
while(pfreq[pfreq_minid] < minfreq):
|
||||
for t in range(len(fft_list)):
|
||||
fft_list[t][pfreq_minid] = 0+0j
|
||||
pfreq_minid += 1
|
||||
|
||||
while(pfreq[pfreq_maxid] > maxfreq):
|
||||
for t in range(len(fft_list)):
|
||||
fft_list[t][pfreq_maxid] = 0+0j
|
||||
pfreq_maxid -= 1
|
||||
|
||||
new_times = []
|
||||
new_freqs = []
|
||||
new_ampls = []
|
||||
new_kept = []
|
||||
|
||||
# i = time, j = freq
|
||||
for i in range(len(fft_list)):
|
||||
#returns a list of couples [id, value]
|
||||
elements = heapq.nlargest(count, enumerate(fft_list[i]), key=lambda x: x[1])
|
||||
|
||||
for idx in range(len(elements)):
|
||||
if(elements[idx][0] < len(pfreq)):
|
||||
new_times.append(offset + i*segsize)
|
||||
new_freqs.append(pfreq[elements[idx][0]])
|
||||
new_ampls.append(fft_list[i][elements[idx][0]])
|
||||
|
||||
# -------------------------------------------- Get amp distribution -------------------------------------------- #
|
||||
new_new_amps = [0 for i in range(int(sample_rate*songlen))]
|
||||
new_new_t = [offset + i/sample_rate for i in range(int(sample_rate*songlen))]
|
||||
|
||||
amp_ct = 0
|
||||
incr_a = segsize*4
|
||||
len_seg_a = int(sample_rate*incr_a)
|
||||
count_a = len_seg_a//1000
|
||||
left_0 = int(sample_rate*(amp_ct+offset))
|
||||
while(amp_ct < songlen-segsize):
|
||||
left = int(sample_rate*(amp_ct+offset))
|
||||
right = int(sample_rate*(amp_ct+offset + incr_a))
|
||||
|
||||
#returns a list of couples [id, value]
|
||||
elements = heapq.nlargest(count_a, enumerate([song_data[i] for i in range(left, right)]), key=lambda x: x[1])
|
||||
|
||||
amp_ct += incr_a
|
||||
|
||||
for idx in range(len(elements)):
|
||||
try:
|
||||
new_new_amps[elements[idx][0]+left-left_0] = song_data[left+elements[idx][0]]
|
||||
except:
|
||||
pass
|
||||
|
||||
mmxx = max(new_new_amps)
|
||||
new_new_amps = [nnw*1000/mmxx for nnw in new_new_amps]
|
||||
|
||||
# localize peaks
|
||||
left_id = 0
|
||||
right_id = 0
|
||||
a_ampl = 0
|
||||
in_seg = False
|
||||
time_d = 0.035
|
||||
cur_t = 0
|
||||
|
||||
last_t = -10.0
|
||||
|
||||
locs = [] # amplitudes
|
||||
loct = [] # times
|
||||
for i in range(len(new_new_amps)):
|
||||
if(new_new_amps[i] > 100):
|
||||
if(not in_seg):
|
||||
in_seg = True
|
||||
left_id = i
|
||||
right_id = i
|
||||
a_ampl = max(a_ampl, new_new_amps[i])
|
||||
cur_t = 0
|
||||
else:
|
||||
cur_t += 1/sample_rate
|
||||
if(in_seg and cur_t >= time_d):
|
||||
in_seg = False
|
||||
delta_t = (right_id - left_id)/sample_rate
|
||||
if(np.abs(left_id/sample_rate - last_t) >= 0.01): # these notes are less than 10ms apart !
|
||||
last_t = right_id/sample_rate
|
||||
if(delta_t < segsize*1.1):
|
||||
locs.append(a_ampl)
|
||||
loct.append((left_id + right_id)/(2*sample_rate) + offset)
|
||||
else:
|
||||
locs.append(a_ampl)
|
||||
loct.append([left_id/sample_rate + offset, right_id/sample_rate + offset])
|
||||
|
||||
a_ampl = 0
|
||||
|
||||
# -------------------------------------------- Compute freqs -------------------------------------------- #
|
||||
|
||||
ssize_0 = segsize/3
|
||||
locf = [] # frequencies
|
||||
for k in range(len(locs)):
|
||||
ktime = 0
|
||||
ssize = ssize_0
|
||||
if(type(loct[k]) == float): # circle
|
||||
ktime = loct[k]
|
||||
else: # slider
|
||||
ktime = (loct[k][1]+loct[k][0])/2
|
||||
ssize = max((loct[k][1]-loct[k][0])/2, ssize_0)
|
||||
|
||||
left_id = max(0, int((ktime-ssize/2)*sample_rate))
|
||||
|
||||
right_id = min(int((ktime+ssize/2)*sample_rate), len(song_data))
|
||||
|
||||
# calculate the fft
|
||||
pff = scp.fft.rfft(song_data[left_id:right_id])
|
||||
|
||||
fmax = pfreq[0]
|
||||
fampmax = 0
|
||||
for i in range(1, len(pff)):
|
||||
if(pfreq[i] > minfreq and pfreq[i] < maxfreq and fampmax < np.abs(pff[i])):
|
||||
fmax = pfreq[i]
|
||||
fampmax = np.abs(pff[i])
|
||||
|
||||
locf.append(fmax)
|
||||
|
||||
# -------------------------------------------- Merge -------------------------------------------- #
|
||||
|
||||
k = 0
|
||||
while(k < len(locs)):
|
||||
delta_t = 0
|
||||
if(type(loct[k]) == float):
|
||||
delta_t += loct[k]
|
||||
else:
|
||||
delta_t += (loct[k][0] + loct[k][1])/2
|
||||
|
||||
if(type(loct[k-1]) == float):
|
||||
delta_t -= loct[k-1]
|
||||
else:
|
||||
delta_t -= (loct[k-1][0] + loct[k-1][1])/2
|
||||
if(k > 0 and np.abs(delta_t) < segsize and np.abs(locs[k] - locs[k-1]) < 50 and is_note_within(locf[k], locf[k-1])):
|
||||
loct[k-1] = [loct[k-1], loct[k]]
|
||||
locs[k-1] = (locs[k-1] + locs[k])/2
|
||||
loct[k] = -1
|
||||
locs[k] = -1
|
||||
locf[k] = -1
|
||||
loct.remove(-1)
|
||||
locs.remove(-1)
|
||||
locf.remove(-1)
|
||||
k += 1
|
||||
|
||||
|
||||
# -------------------------------------------- Plot -------------------------------------------- #
|
||||
|
||||
if(canPlot):
|
||||
plt_loct_all = []
|
||||
plt_loct = []
|
||||
plt_locs = []
|
||||
plt_slidt = []
|
||||
plt_slids = []
|
||||
for i in range(len(loct)):
|
||||
if(type(loct[i]) == float):
|
||||
plt_loct_all.append(loct[i])
|
||||
plt_loct.append(loct[i])
|
||||
plt_locs.append(locs[i])
|
||||
else:
|
||||
plt_loct_all.append(loct[i][0])
|
||||
plt_slidt.append(loct[i][0])
|
||||
plt_slidt.append(loct[i][1])
|
||||
plt_slids.append(locs[i])
|
||||
plt_slids.append(locs[i])
|
||||
|
||||
plt.plot(new_new_t, new_new_amps, "y-", label="amplitude (ua)")
|
||||
plt.plot(plt_loct, plt_locs, "ro", label="circles")
|
||||
plt.plot(plt_slidt, plt_slids, "go", label="sliders")
|
||||
plt.plot(plt_loct_all, locf, "mo", label="frequencies (Hz)")
|
||||
plt.legend(loc="upper left")
|
||||
|
||||
'''plt.plot(new_times, new_freqs)
|
||||
plt.plot(new_times, [elt*1000/mx for elt in new_ampls])
|
||||
plt.plot(new_times, new_kept, "bo")'''
|
||||
plt.grid()
|
||||
plt.show()
|
||||
|
||||
# -------------------------------------------- Write -------------------------------------------- #
|
||||
|
||||
if(writeO):
|
||||
f = open("result_bad_apple[90].txt", "w")
|
||||
f.write("Song name : " + song_name + "\n")
|
||||
f.write("Start : " + str(offset) + "\n")
|
||||
f.write("End : " + str(offset+songlen) + "\n\n")
|
||||
|
||||
f.write("Hit Objects : \n")
|
||||
for ct in loct:
|
||||
f.write(str(ct))
|
||||
f.write("\n")
|
||||
|
||||
f.close()
|
||||
|
||||
return (loct, locs)
|
||||
|
||||
def convert_to_wav(song_name:str, output_file="audio.wav") -> str:
|
||||
"""
|
||||
Converts the song to .wav, only if it's not already in wave format.
|
||||
Currently relies on file extension.
|
||||
Returns: the song_name that should be used afterwards.
|
||||
"""
|
||||
extension = Path(song_name).suffix
|
||||
if(extension == ".mp3" or extension == ".ogg"):
|
||||
print("Converting to .wav...")
|
||||
subprocess.run(["ffmpeg", "-y", "-i", song_name, output_file], shell=False)
|
||||
return output_file
|
||||
return song_name
|
||||
|
||||
'''
|
||||
# c-type
|
||||
SONG_LEN = 7
|
||||
OFFSET = 0.042
|
||||
BPM = 149.3
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
wavved_song = convert_to_wav("songs/ctype.mp3")
|
||||
'''
|
||||
'''
|
||||
# tetris_2
|
||||
SONG_LEN = 14
|
||||
OFFSET = 0
|
||||
BPM = 157
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
wavved_song = convert_to_wav("songs/tetris_2.wav")
|
||||
'''
|
||||
'''
|
||||
# test
|
||||
SONG_LEN = 1
|
||||
OFFSET = 0
|
||||
BPM = 240
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
'''
|
||||
'''
|
||||
# gmtn
|
||||
SONG_LEN = 5
|
||||
OFFSET = 1.652
|
||||
BPM = 155
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
wavved_song = convert_to_wav("songs/furioso melodia.mp3")
|
||||
'''
|
||||
'''
|
||||
# E
|
||||
SONG_LEN = 15
|
||||
OFFSET = 2.641
|
||||
BPM = 155
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
wavved_song = convert_to_wav("songs/rushe.mp3")
|
||||
'''
|
||||
'''
|
||||
# Tsubaki
|
||||
SONG_LEN = 20
|
||||
OFFSET = 35.659
|
||||
BPM = 199
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
wavved_song = convert_to_wav("songs/TSUBAKI.mp3")
|
||||
'''
|
||||
'''
|
||||
# Owen 1/2
|
||||
SONG_LEN = 20
|
||||
OFFSET = 1.008
|
||||
BPM = 157
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
wavved_song = convert_to_wav("songs/owen(157.00024-1008).mp3")
|
||||
'''
|
||||
'''
|
||||
# Owen 2/2
|
||||
SONG_LEN = 7
|
||||
OFFSET = 25.466
|
||||
BPM = 157
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
wavved_song = convert_to_wav("songs/owen(157.00024-1008).mp3")
|
||||
'''
|
||||
|
||||
# death
|
||||
SONG_LEN = 10
|
||||
OFFSET = 21.750
|
||||
BPM = 180
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
wavved_song = convert_to_wav("songs/Night of Knights.mp3")
|
||||
|
||||
'''
|
||||
# Bad apple
|
||||
SONG_LEN = 15
|
||||
OFFSET = 0.152
|
||||
BPM = 138
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
#wavved_song = convert_to_wav("songs/Bad apple (138-152).mp3")
|
||||
wavved_song = convert_to_wav("songs/Bad apple (138-152)[filtered].wav")
|
||||
'''
|
||||
'''
|
||||
# Freedom dive
|
||||
SONG_LEN = 7
|
||||
OFFSET = 1.058
|
||||
BPM = 222.22
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
#wavved_song = convert_to_wav("songs/Freedom Dive (222.22-1058).mp3")
|
||||
wavved_song = convert_to_wav("songs/Freedom Dive (222.22-1058)[filtered].wav")
|
||||
'''
|
||||
'''
|
||||
# Mevalogania
|
||||
SONG_LEN = 7
|
||||
OFFSET = 7.984
|
||||
BPM = 240
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
#wavved_song = convresult_bad_appleert_to_wav("songs/Megalovania(240-7984).mp3")
|
||||
wavved_song = convert_to_wav("songs/Megalovania(240-7984)[filtered].wav")
|
||||
'''
|
||||
'''
|
||||
SONG_LEN = 0 # length of the song, in seconds
|
||||
OFFSET = 0 # offset of the 1st note (aka time offset of the first red bar), in seconds
|
||||
BPM = 0 # BPM
|
||||
wavved_song = convert_to_wav("insert_song_name_here.wav")
|
||||
'''
|
||||
# Do not touch
|
||||
DIVIDER = 4 # note divider
|
||||
SEGSIZE = 1/(BPM/60)
|
||||
NOTE_DIST = (2**(1/4))
|
||||
OCTAVE_DIST = 0.05
|
||||
|
||||
# keep_highest(song_name, offset, songlen, segsize, count, output_name, minfreq=110, maxfreq=5000, ampthr=250):
|
||||
(loct, locs) = keep_highest(wavved_song, OFFSET, SONG_LEN, SEGSIZE/DIVIDER, 1, "Zblit.wav", minfreq=220, maxfreq=3000, ampthr=800)
|
||||
|
||||
'''
|
||||
minfreq and maxfred are thresholds for frequency analysts (anything outside of [minfreq, maxfreq] will not be accounted for)
|
||||
ampthr is a threshold for amplitude (arbitrary unit)
|
||||
'''
|
||||
|
||||
''' you can deactivate this if you want (show timings points in terminal) '''
|
||||
'''
|
||||
import time
|
||||
import random
|
||||
loct2 = []
|
||||
|
||||
for k in loct:
|
||||
if(type(k) == float):
|
||||
loct2.append(k)
|
||||
else:
|
||||
loct2.append(k[0])
|
||||
loct2.append(k[1])
|
||||
|
||||
for i in range(len(loct2)-1):
|
||||
print("*"*(random.randint(10, 100)))
|
||||
time.sleep(loct2[i+1]-loct2[i])
|
||||
|
||||
print("yipee")
|
||||
'''
|
||||
# complexity test
|
||||
fl = open("complexity.txt", "w")
|
||||
|
||||
# f.write("Song name : " + song_name + "\n")
|
||||
'''
|
||||
deltat = []
|
||||
compl = []
|
||||
for end in range(2,120):
|
||||
st = datetime.datetime.now()
|
||||
(e, ee) = keep_highest(wavved_song, OFFSET, OFFSET+end/2, SEGSIZE/DIVIDER, 1, "Zblit.wav", minfreq=220, maxfreq=3000, ampthr=800, canPlot=False,writeO=False)
|
||||
et = datetime.datetime.now()
|
||||
dt = et.microsecond - st.microsecond + (et.second - st.second)*1000000 + (et.minute - st.minute)/60
|
||||
if(dt>0):
|
||||
deltat.append(end/2)
|
||||
compl.append(dt)
|
||||
|
||||
plt.plot(deltat, compl, "y-")
|
||||
plt.plot(deltat, compl, "ro")
|
||||
plt.xlabel("size of the song")
|
||||
plt.ylabel("time complexity (us)")
|
||||
|
||||
plt.grid()
|
||||
plt.show()
|
||||
|
||||
fl.close()
|
||||
'''
|
||||
Binary file not shown.
|
After Width: | Height: | Size: 97 KiB |
|
|
@ -0,0 +1,183 @@
|
|||
# treat_as_circle
|
||||
# -------------------
|
||||
def treat_as_circle(ind, is_slider_head):
|
||||
global timeM1
|
||||
start = (ind == 0)
|
||||
beat_phase = round((objects[ind].time.total_seconds()*1000 - offset*1000)/ms_per_beat % 1, 2)
|
||||
xcoord, ycoord = objects[ind].position
|
||||
xcoord, ycoord = normalize(xcoord, ycoord)
|
||||
if start:
|
||||
delta_time = 0
|
||||
else:
|
||||
delta_time = (objects[ind].time - timeM1).total_seconds()
|
||||
red_tick = (beat_phase == 0.5)
|
||||
blue_tick = (beat_phase == 0.75)
|
||||
white_tick = (beat_phase == 1.0)
|
||||
delta_beat = round((delta_time*1000)/ms_per_beat, 2) / 3
|
||||
timeM1 = objects[ind].time
|
||||
stream.append((start,delta_time,delta_beat,blue_tick,red_tick,white_tick,xcoord,ycoord,False,is_slider_head, stars))
|
||||
|
||||
|
||||
|
||||
# Création de "stream"
|
||||
#-------------------------------
|
||||
def extract_stream(map:sl.beatmap.Beatmap, fix_stars=-1):
|
||||
if fix_stars == -1:
|
||||
stars = map.stars() / 7
|
||||
else:
|
||||
stars = fix_stars / 7
|
||||
objects = map.hit_objects()
|
||||
stream = []
|
||||
offset = map.timing_points[0].offset.total_seconds()
|
||||
ms_per_beat = map.timing_points[0].ms_per_beat
|
||||
timeM1 = timedelta(seconds=0)
|
||||
n = len(objects)
|
||||
for i in range(len(objects)):
|
||||
if is_slider(objects[i]):
|
||||
duration = objects[i].end_time - objects[i].time
|
||||
positions = []
|
||||
number_points,curve = len(objects[i].curve.points)
|
||||
curve = objects[i].curve.points
|
||||
delta = duration/float(number_points)
|
||||
for j in range(number_points):
|
||||
if j == 0:
|
||||
treat_as_circle(i, True)
|
||||
else:
|
||||
start = False
|
||||
xcoord, ycoord = normalize(curve[j][0], curve[j][1])
|
||||
delta_time = delta.total_seconds()
|
||||
beat_phase = round(((objects[i].time + j*delta).total_seconds()*1000 - offset*1000)/ms_per_beat % 1, 2)
|
||||
delta_beat = round((delta_time*1000)/ms_per_beat, 2) / 3
|
||||
red_tick = (beat_phase == 0.5)
|
||||
blue_tick = (beat_phase == 0.75)
|
||||
white_tick = (beat_phase == 1.0)
|
||||
stream.append((start,delta_time,delta_beat,blue_tick,red_tick,white_tick,xcoord,ycoord,True,False, stars))
|
||||
timeM1 = objects[i].end_time
|
||||
else:
|
||||
treat_as_circle(i, False)
|
||||
return stream
|
||||
|
||||
|
||||
|
||||
#Construction de séquence 2
|
||||
# --------------------------------------
|
||||
def mirror_stream(stream):
|
||||
return [(s[0], s[1], s[2], s[3], s[4], s[5], 1 - s[6], s[7], s[8], s[9], s[10]) for s in stream]
|
||||
|
||||
def build_sequences(maps:list):
|
||||
sequences = []
|
||||
for mapp in maps:
|
||||
stream = extract_stream(mapp)
|
||||
mirror = mirror_stream(stream)
|
||||
for i in range(len(stream) - seq_len-1):
|
||||
sequences.append((stream[i:i+seq_len], stream[i+seq_len+1]))
|
||||
sequences.append((mirror[i:i+seq_len], mirror[i+seq_len+1]))
|
||||
return sequences
|
||||
|
||||
|
||||
#Création de map
|
||||
# --------------------------------------
|
||||
def reposition_objects(mapp: sl.Beatmap):
|
||||
stream = extract_stream(mapp, fix_stars=5.5)
|
||||
model.eval()
|
||||
j = 0 # because stream separates different curve points, j keeps the true object index while str_i is the index corresponding to "fake" objects
|
||||
str_i = 0
|
||||
objects = mapp.hit_objects()
|
||||
while str_i < len(stream):
|
||||
with torch.no_grad():
|
||||
if not is_slider(objects[j]):
|
||||
in_sequence = torch.tensor(stream[max(0, str_i - 20):str_i+1], dtype=torch.float32).unsqueeze(0).to(cuda)
|
||||
coords = model(in_sequence)
|
||||
coords = coords.tolist()
|
||||
x,y = coords[0][0], coords[0][1]
|
||||
mapp.hit_objects()[j].position = sl.Position(x*512, y*384)
|
||||
stream[str_i] = [stream[str_i][0], stream[str_i][1], stream[str_i][2], stream[str_i][3], stream[str_i][4], stream[str_i][5], x, y, stream[str_i][8], stream[str_i][9], stream[str_i][10]]
|
||||
j += 1
|
||||
str_i += 1
|
||||
else:
|
||||
for curvepoint_k in range(len(objects[j].curve.points)):
|
||||
in_sequence = torch.tensor(stream[max(0, str_i - 16):str_i+1], dtype=torch.float32).unsqueeze(0).to(cuda)
|
||||
coords = model(in_sequence) # Predict xcoord and ycoord
|
||||
coords = coords.tolist()
|
||||
x,y = coords[0][0], coords[0][1]
|
||||
if curvepoint_k == 0:
|
||||
mapp.hit_objects()[j].position = sl.Position(x*512, y*384)
|
||||
mapp.hit_objects()[j].curve.points[curvepoint_k] = sl.Position(x*512, y*384)
|
||||
stream[str_i] = [stream[str_i][0], stream[str_i][1], stream[str_i][2], stream[str_i][3], stream[str_i][4], stream[str_i][5], x, y, stream[str_i][8], stream[str_i][9], stream[str_i][10]]
|
||||
str_i += 1
|
||||
j += 1
|
||||
|
||||
return mapp
|
||||
path = Path("/kaggle/input/survey1/4.osu")
|
||||
map = reposition_objects(sl.Beatmap.from_path(path))
|
||||
map.write_path("/kaggle/working/4.osu")
|
||||
|
||||
|
||||
#TRAINING LOOP
|
||||
# --------------------------------------
|
||||
model = LSTM(input_size=num_features, hidden_size=hidden_size, output_size=2).to(cuda)
|
||||
optimizer = Adam(model.parameters(), lr=0.003)
|
||||
criterion = nn.L1Loss()
|
||||
# boucle d'entraînement
|
||||
for epoch in range(20):
|
||||
model.train()
|
||||
train_loss = 0
|
||||
for x, y in train_dataloader:
|
||||
optimizer.zero_grad()
|
||||
preds = model(x)
|
||||
loss = criterion(preds, y)
|
||||
loss.backward()
|
||||
optimizer.step()
|
||||
train_loss += loss.item()
|
||||
avg_train_loss = train_loss / len(train_dataloader)
|
||||
model.eval()
|
||||
val_loss = 0
|
||||
with torch.no_grad():
|
||||
for x_val, y_val in val_dataloader:
|
||||
preds_val = model(x_val)
|
||||
val_loss += criterion(preds_val, y_val).item()
|
||||
|
||||
avg_val_loss = val_loss / len(val_dataloader)
|
||||
print(f'Epoch {epoch+1}: Train Loss = {avg_train_loss:.4f}, Val Loss = {avg_val_loss:.4f}')
|
||||
|
||||
#CREATION DATASET
|
||||
# --------------------------------------
|
||||
class SequenceDataset(torch.utils.data.Dataset):
|
||||
def __init__(self, sequences):
|
||||
self.sequences = sequences # liste de tuples
|
||||
|
||||
def __len__(self):
|
||||
return len(self.sequences)
|
||||
|
||||
def __getitem__(self, idx):
|
||||
x, y = self.sequences[idx]
|
||||
x = torch.tensor(x, dtype=torch.float32).to(cuda) # x = entrée
|
||||
y = torch.tensor([y[6], y[7]], dtype=torch.float32).to(cuda) # y = sortie ([xcoord, ycoord])
|
||||
return x, y
|
||||
path1 = Path("/kaggle/input/biggerset139")
|
||||
path2 = Path("/kaggle/input/varietypack51")
|
||||
maps = [sl.Beatmap.from_path(m) for m in (path1.ls() + path2.ls())]
|
||||
sequences = build_sequences(maps)
|
||||
train_sequences, val_sequences = train_test_split(sequences, test_size=0.2)
|
||||
train_dataset = SequenceDataset(train_sequences)
|
||||
val_dataset = SequenceDataset(val_sequences)
|
||||
train_dataloader = DataLoader(train_dataset, batch_size=64, shuffle=True)
|
||||
val_dataloader = DataLoader(val_dataset, batch_size=64, shuffle=False)
|
||||
|
||||
|
||||
|
||||
|
||||
#DEFINITION LSTM/dataset
|
||||
# --------------------------------------
|
||||
hidden_size,num_features,num_outputs = 128,11,2
|
||||
class LSTM(nn.Module):
|
||||
def __init__(self, input_size, hidden_size, output_size):
|
||||
super().__init__()
|
||||
self.lstm = nn.LSTM(input_size, hidden_size, batch_first=True)
|
||||
self.fc = nn.Linear(hidden_size, output_size)
|
||||
|
||||
def forward(self, x):
|
||||
out, _ = self.lstm(x)
|
||||
out = self.fc(out[:, -1, :])
|
||||
out = torch.sigmoid(out)
|
||||
return out
|
||||
Binary file not shown.
|
After Width: | Height: | Size: 156 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 132 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 94 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 71 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 109 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 91 KiB |
|
|
@ -0,0 +1 @@
|
|||
pandoc presentation.md -t beamer -o presentation.pdf
|
||||
|
|
@ -0,0 +1,178 @@
|
|||
---
|
||||
title: "Construction automatique de niveaux pour le jeu de rythme osu! à partir de fichiers audio"
|
||||
topic: "osu"
|
||||
author: "Aboujaib Alexandre - 28173"
|
||||
theme: "Copenhagen"
|
||||
header-includes:
|
||||
- \AtBeginDocument{\title[Construction automatique de niveaux pour osu!]{Construction automatique de niveaux pour le jeu de rythme osu! à partir de fichiers audio}}
|
||||
---
|
||||
|
||||
# Presentation du problème
|
||||
|
||||
### Le jeu osu!
|
||||
|
||||
- Jeu de rythme pour PC
|
||||
- Sorti le 16 septembre 2007, adapté d'un jeu pour DS
|
||||
- Environ 500 000 joueurs actifs
|
||||
- Se joue avec une souris/tablette et clavier
|
||||
{ width=50% }
|
||||
{ width=40% }
|
||||
|
||||
### Les beatmaps (niveaux)
|
||||
::: columns
|
||||
:::: column
|
||||
{ width=110% }
|
||||
|
||||
::::
|
||||
:::: column
|
||||
{ width=110% }
|
||||
::::
|
||||
|
||||
:::
|
||||
|
||||
### Exemples de cercles
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
### Exemples de sliders
|
||||
|
||||
{ width=50% }
|
||||
{ width=50% }
|
||||
|
||||
|
||||
|
||||
|
||||
### Formulation du problème
|
||||
|
||||
Nous nous proposons de **créer deux programmes** permettant **au mieux**, à partir d’un **fichier
|
||||
audio** donné, de **construire un niveau pour osu!**.
|
||||
|
||||
|
||||
# Partie I : Analyse de musique
|
||||
|
||||
### Approche en deux temps
|
||||
|
||||
#### Spécifications
|
||||
|
||||
*Entrée :* un fichier audio (format quelconque)
|
||||
|
||||
*Sortie :* des données relatives à la musique permettant le placement des notes :
|
||||
|
||||
`(double | (double * double)) list`
|
||||
|
||||
#### Processus retenu ici
|
||||
|
||||
|
||||
|
||||
### Schéma du processus
|
||||
|
||||
|
||||
|
||||
### Filtres physiques & Transformée de Fourier
|
||||
|
||||
::: columns
|
||||
:::: column
|
||||
{ width=90% }
|
||||
::::
|
||||
:::: column
|
||||
|
||||
::::
|
||||
:::
|
||||
|
||||
### Résultats
|
||||
|
||||
{ width=60% }
|
||||
|
||||
Résultat de l'extraction sur une musique (Bad Apple) pendant 15s
|
||||
|
||||
### Résultats
|
||||
|
||||
|
||||
|
||||
Résultat de l'extraction sur la même musique pendant 120s
|
||||
|
||||
### Résultats
|
||||
|
||||
#### Complexité
|
||||
|
||||
{ width=95% }
|
||||
|
||||
### Limites, saturation et améliorations
|
||||
|
||||
|
||||
|
||||
### Limites, saturation et améliorations
|
||||
|
||||
|
||||
|
||||
> les fréquences ne sont pas correctement détéctées ici (car trop d'harmoniques)
|
||||
|
||||
### Limites, saturation et améliorations
|
||||
|
||||
|
||||
|
||||
### Limites, saturation et améliorations
|
||||
|
||||
|
||||
|
||||
> une musique "dense" fait que la méthode d'extraction des amplitudes n'est pas précise
|
||||
|
||||
# Partie II : placement spatial
|
||||
|
||||
# Annexe : code Python
|
||||
|
||||
### librairies utilisées ici
|
||||
|
||||
{ width=70% }
|
||||
|
||||
### quelques fonctions utiles
|
||||
|
||||
|
||||
|
||||
### `parse music` : extraction de la liste des amplitudes et du "sample rate"
|
||||
|
||||
::: columns
|
||||
:::: column
|
||||
{ width=80% }
|
||||
::::
|
||||
:::: column
|
||||
|
||||
::::
|
||||
:::
|
||||
|
||||
### `cleaning` : retirer les fréquences trop basses et hautes
|
||||
|
||||
{ width=60% }
|
||||
|
||||
### `get amp distribution` : usage de files de priorité pour extraire les maxima
|
||||
|
||||
{ width=50% }
|
||||
|
||||
### `get frequency distribution` : fréquence maximale pour chaque point
|
||||
|
||||
{ width=60% }
|
||||
|
||||
### `get sliders` : fusionner les notes trop proches en des 'sliders'
|
||||
|
||||
{ width=75% }
|
||||
|
||||
### `draw` : afficher les résultats
|
||||
|
||||
{ width=60% }
|
||||
|
||||
### quelques données de test
|
||||
|
||||
::: columns
|
||||
:::: column
|
||||
{ width=75% }
|
||||
::::
|
||||
:::: column
|
||||
|
||||
::::
|
||||
:::
|
||||
|
||||
### interface
|
||||
|
||||
|
||||
Binary file not shown.
|
|
@ -0,0 +1,192 @@
|
|||
\hypertarget{presentation-du-probluxe8me}{%
|
||||
\section{Presentation du problème}\label{presentation-du-probluxe8me}}
|
||||
|
||||
\hypertarget{pruxe9sentation-dune-beatmap}{%
|
||||
\subsection{Présentation d'une
|
||||
beatmap}\label{pruxe9sentation-dune-beatmap}}
|
||||
|
||||
\begin{frame}{Le Jeu osu!}
|
||||
\protect\hypertarget{le-jeu-osu}{}
|
||||
\begin{itemize}
|
||||
\tightlist
|
||||
\item
|
||||
Jeu de rythme pour PC
|
||||
\item
|
||||
Sorti le 16 septembre 2007
|
||||
\item
|
||||
Environ 500 000 joueurs actifs
|
||||
\item
|
||||
Se joue avec une souris/tablette et clavier {[}osuLogo.png{]}
|
||||
{[}tablet.png{]}
|
||||
\end{itemize}
|
||||
\end{frame}
|
||||
|
||||
\begin{frame}{les beatmaps}
|
||||
\protect\hypertarget{les-beatmaps}{}
|
||||
\begin{block}{3 types d'objets:}
|
||||
\protect\hypertarget{types-dobjets}{}
|
||||
\begin{itemize}
|
||||
\item
|
||||
cercles
|
||||
\item
|
||||
sliders
|
||||
\item
|
||||
spinners (ignorés)
|
||||
\item
|
||||
Lien musique \textless-\textgreater{} niveau
|
||||
\item
|
||||
Difficulté quantifiée par des ``étoiles''
|
||||
\end{itemize}
|
||||
\end{block}
|
||||
\end{frame}
|
||||
|
||||
\begin{frame}{Formulation du problème}
|
||||
\protect\hypertarget{formulation-du-probluxe8me}{}
|
||||
Nous nous proposons de créer deux programmes permettant au mieux, à
|
||||
partir d'un fichier audio donné, de construire un niveau pour osu!.
|
||||
\end{frame}
|
||||
|
||||
\hypertarget{duxe9composition-du-probluxe8me}{%
|
||||
\subsection{Décomposition du
|
||||
problème}\label{duxe9composition-du-probluxe8me}}
|
||||
|
||||
\begin{frame}{Analyse de musique (Alexandre)}
|
||||
\protect\hypertarget{analyse-de-musique-alexandre}{}
|
||||
\begin{itemize}
|
||||
\tightlist
|
||||
\item
|
||||
respecter les caractéristiques de la musique : décalage, tempo
|
||||
\item
|
||||
distinguer sons importants et accessoires
|
||||
\end{itemize}
|
||||
\end{frame}
|
||||
|
||||
\begin{frame}{Placement des objets}
|
||||
\protect\hypertarget{placement-des-objets}{}
|
||||
\begin{itemize}
|
||||
\tightlist
|
||||
\item
|
||||
angles -\textgreater{} \emph{flow}
|
||||
\item
|
||||
doit complémenter la musique (espacement, difficulté)
|
||||
\item
|
||||
visuel: plaisant et lisible (-\textgreater{} motifs/patterns)
|
||||
\end{itemize}
|
||||
\end{frame}
|
||||
|
||||
\hypertarget{ruxe9duction-placement-pruxe9diction-de-suxe9quence}{%
|
||||
\subsection{Réduction: placement \textless{} prédiction de
|
||||
séquence}\label{ruxe9duction-placement-pruxe9diction-de-suxe9quence}}
|
||||
|
||||
\begin{frame}{Observation}
|
||||
\protect\hypertarget{observation}{}
|
||||
Le placement du prochain objet dépend:
|
||||
|
||||
\begin{itemize}
|
||||
\tightlist
|
||||
\item
|
||||
de la musique
|
||||
\item
|
||||
des objets précédents
|
||||
\end{itemize}
|
||||
|
||||
\textasciitilde{} prédiction de séquence (en économie par ex.)
|
||||
|
||||
\textbf{BUT} = encoder musique et objets pour prédire les prochains
|
||||
objets
|
||||
\end{frame}
|
||||
|
||||
\begin{frame}{1ère approche: glouton}
|
||||
\protect\hypertarget{uxe8re-approche-glouton}{}
|
||||
\begin{block}{Un principe simple\ldots{}}
|
||||
\protect\hypertarget{un-principe-simple}{}
|
||||
\begin{itemize}
|
||||
\tightlist
|
||||
\item
|
||||
à partir des choix précédents -\textgreater{} faire le meilleur choix
|
||||
\end{itemize}
|
||||
\end{block}
|
||||
|
||||
\begin{block}{\ldots mais peu efficace en pratique.}
|
||||
\protect\hypertarget{mais-peu-efficace-en-pratique.}{}
|
||||
\begin{itemize}
|
||||
\tightlist
|
||||
\item
|
||||
beaucoup d'heuristiques à coder -\textgreater{} compliqué sur un jeu
|
||||
comme osu!
|
||||
\end{itemize}
|
||||
\end{block}
|
||||
\end{frame}
|
||||
|
||||
\begin{frame}{Réseaux de neurones nécurrents}
|
||||
\protect\hypertarget{ruxe9seaux-de-neurones-nuxe9currents}{}
|
||||
{[}{]}
|
||||
\end{frame}
|
||||
|
||||
\begin{frame}{Amélioration: LSTM}
|
||||
\protect\hypertarget{amuxe9lioration-lstm}{}
|
||||
{[}{]}
|
||||
\end{frame}
|
||||
|
||||
\begin{frame}{Encodage précis des objets}
|
||||
\protect\hypertarget{encodage-pruxe9cis-des-objets}{}
|
||||
\begin{block}{Importance de l'encodage}
|
||||
\protect\hypertarget{importance-de-lencodage}{}
|
||||
Permet au modèle ``d'apprendre'' le plus efficacement possible
|
||||
\end{block}
|
||||
|
||||
\begin{block}{Encodage retenu}
|
||||
\protect\hypertarget{encodage-retenu}{}
|
||||
10-uplet contenant: start, delta\_time (en ms), delta\_beat (float),
|
||||
blue\_tick (bool), red\_tick (bool), white\_tick (bool), xcoord (float),
|
||||
ycoord (float), is\_curve\_point, is\_slider\_start
|
||||
|
||||
Normalisation des données
|
||||
\end{block}
|
||||
\end{frame}
|
||||
|
||||
\begin{frame}{Protocole d'entraînement}
|
||||
\protect\hypertarget{protocole-dentrauxeenement}{}
|
||||
\begin{itemize}
|
||||
\tightlist
|
||||
\item
|
||||
Echantillon de 188 niveaux, style
|
||||
\end{itemize}
|
||||
\end{frame}
|
||||
|
||||
\hypertarget{ruxe9sultats}{%
|
||||
\subsection{Résultats}\label{ruxe9sultats}}
|
||||
|
||||
\begin{frame}{1ère observations}
|
||||
\protect\hypertarget{uxe8re-observations}{}
|
||||
\begin{block}{Choix de musiques d'essai :}
|
||||
\protect\hypertarget{choix-de-musiques-dessai}{}
|
||||
\begin{itemize}
|
||||
\tightlist
|
||||
\item
|
||||
1 niveau contrôle = humain (Science)
|
||||
\item
|
||||
1 niveau de type ``stream'' (Entanglement)
|
||||
\item
|
||||
1 niveau de type ``jump'' (Rooftop)
|
||||
\item
|
||||
1 niveau hybride (Blade Dance)
|
||||
\end{itemize}
|
||||
|
||||
Rythmes : humains
|
||||
\end{block}
|
||||
\end{frame}
|
||||
|
||||
\begin{frame}{Améliorations possibles?}
|
||||
\protect\hypertarget{amuxe9liorations-possibles}{}
|
||||
\begin{itemize}
|
||||
\tightlist
|
||||
\item
|
||||
autres types de modèles
|
||||
\item
|
||||
plus d'entraînement (mais surapprentissage\ldots)
|
||||
\item
|
||||
meilleur lien musique \textless-\textgreater{} placement (ici mis de
|
||||
côté\ldots)
|
||||
\end{itemize}
|
||||
\end{frame}
|
||||
|
|
@ -0,0 +1,68 @@
|
|||
Song name : audio.wav
|
||||
Start : 21.75
|
||||
End : 31.75
|
||||
|
||||
Hit Objects :
|
||||
21.80657596371882
|
||||
21.86517006802721
|
||||
21.954943310657598
|
||||
22.158684807256236
|
||||
22.288934240362813
|
||||
22.47361678004535
|
||||
22.554920634920634
|
||||
22.686349206349206
|
||||
22.807539682539684
|
||||
22.91013605442177
|
||||
23.045102040816328
|
||||
23.131043083900227
|
||||
23.180090702947847
|
||||
23.220328798185943
|
||||
[23.287528344671202, 23.367142857142856]
|
||||
[23.472698412698414, 23.55392290249433]
|
||||
23.821655328798187
|
||||
23.95299319727891
|
||||
24.15141723356009
|
||||
24.22986394557823
|
||||
24.498219954648526
|
||||
24.66047619047619
|
||||
[24.78424036281179, 24.884920634920636]
|
||||
25.170090702947846
|
||||
25.489603174603175
|
||||
25.57687074829932
|
||||
25.655873015873016
|
||||
25.812653061224488
|
||||
[25.908027210884356, 25.956757369614515]
|
||||
25.993038548752835
|
||||
[26.150385487528347, 26.199002267573697]
|
||||
26.305804988662132
|
||||
26.492925170068027
|
||||
26.652392290249434
|
||||
26.81343537414966
|
||||
27.013990929705216
|
||||
27.16793650793651
|
||||
27.28827664399093
|
||||
27.49448979591837
|
||||
27.838990929705215
|
||||
27.88766439909297
|
||||
27.953197278911563
|
||||
28.156689342403627
|
||||
28.287687074829932
|
||||
28.502426303854875
|
||||
28.686780045351476
|
||||
28.83141723356009
|
||||
29.16842403628118
|
||||
29.285532879818593
|
||||
[29.477437641723355, 29.53560090702948]
|
||||
29.660725623582767
|
||||
29.810306122448978
|
||||
29.866689342403628
|
||||
30.13770975056689
|
||||
[30.484603174603173, 30.555702947845806]
|
||||
30.623378684807257
|
||||
30.807585034013606
|
||||
30.956666666666667
|
||||
31.155680272108846
|
||||
31.289024943310658
|
||||
31.38056689342404
|
||||
31.4947052154195
|
||||
31.62167800453515
|
||||
|
|
@ -0,0 +1,63 @@
|
|||
Song name : songs/Bad apple (138-152)[filtered].wav
|
||||
Start : 0.152
|
||||
End : 20.152
|
||||
|
||||
Hit Objects :
|
||||
0.17748752834467119
|
||||
0.6138367346938776
|
||||
1.0485873015873015
|
||||
1.4824761904761903
|
||||
[1.5911723356009069, 1.6998571428571427]
|
||||
[1.80856462585034, 1.916580498866213]
|
||||
2.352963718820862
|
||||
2.786183673469388
|
||||
3.2216031746031746
|
||||
3.4390068027210887
|
||||
3.6557414965986395
|
||||
4.092079365079365
|
||||
4.525321995464853
|
||||
4.960752834467121
|
||||
5.0694263038548755
|
||||
[5.178133786848073, 5.286818594104308]
|
||||
5.394857142857143
|
||||
5.8312176870748305
|
||||
6.264437641723356
|
||||
6.699879818594105
|
||||
6.917260770975057
|
||||
7.133995464852608
|
||||
7.570333333333333
|
||||
8.00357596371882
|
||||
8.439006802721087
|
||||
8.547680272108844
|
||||
[8.65639909297052, 8.765083900226756]
|
||||
8.873122448979592
|
||||
9.309471655328798
|
||||
9.742714285714285
|
||||
10.178133786848072
|
||||
10.395537414965986
|
||||
10.612272108843536
|
||||
11.049222222222221
|
||||
11.481841269841269
|
||||
11.917272108843537
|
||||
12.025945578231292
|
||||
12.134641723356008
|
||||
12.243371882086167
|
||||
12.35137641723356
|
||||
12.788360544217687
|
||||
13.223145124716552
|
||||
13.655741496598639
|
||||
14.090480725623582
|
||||
14.526897959183673
|
||||
14.960072562358276
|
||||
15.394879818594104
|
||||
15.829607709750567
|
||||
16.264414965986397
|
||||
16.69919954648526
|
||||
17.13463038548753
|
||||
17.352011337868483
|
||||
17.568780045351474
|
||||
18.00577551020408
|
||||
18.44058276643991
|
||||
18.874732426303854
|
||||
19.30789569160998
|
||||
19.744233560090706
|
||||
|
|
@ -0,0 +1,68 @@
|
|||
Song name : audio.wav
|
||||
Start : 21.75
|
||||
End : 31.75
|
||||
|
||||
Hit Objects :
|
||||
21.80657596371882
|
||||
21.86517006802721
|
||||
21.954943310657598
|
||||
22.158684807256236
|
||||
22.288934240362813
|
||||
22.47361678004535
|
||||
22.554920634920634
|
||||
22.686349206349206
|
||||
22.807539682539684
|
||||
22.91013605442177
|
||||
23.045102040816328
|
||||
23.131043083900227
|
||||
23.180090702947847
|
||||
23.220328798185943
|
||||
[23.287528344671202, 23.367142857142856]
|
||||
[23.472698412698414, 23.55392290249433]
|
||||
23.821655328798187
|
||||
23.95299319727891
|
||||
24.15141723356009
|
||||
24.22986394557823
|
||||
24.498219954648526
|
||||
24.66047619047619
|
||||
[24.78424036281179, 24.884920634920636]
|
||||
25.170090702947846
|
||||
25.489603174603175
|
||||
25.57687074829932
|
||||
25.655873015873016
|
||||
25.812653061224488
|
||||
[25.908027210884356, 25.956757369614515]
|
||||
25.993038548752835
|
||||
[26.150385487528347, 26.199002267573697]
|
||||
26.305804988662132
|
||||
26.492925170068027
|
||||
26.652392290249434
|
||||
26.81343537414966
|
||||
27.013990929705216
|
||||
27.16793650793651
|
||||
27.28827664399093
|
||||
27.49448979591837
|
||||
27.838990929705215
|
||||
27.88766439909297
|
||||
27.953197278911563
|
||||
28.156689342403627
|
||||
28.287687074829932
|
||||
28.502426303854875
|
||||
28.686780045351476
|
||||
28.83141723356009
|
||||
29.16842403628118
|
||||
29.285532879818593
|
||||
[29.477437641723355, 29.53560090702948]
|
||||
29.660725623582767
|
||||
29.810306122448978
|
||||
29.866689342403628
|
||||
30.13770975056689
|
||||
[30.484603174603173, 30.555702947845806]
|
||||
30.623378684807257
|
||||
30.807585034013606
|
||||
30.956666666666667
|
||||
31.155680272108846
|
||||
31.289024943310658
|
||||
31.38056689342404
|
||||
31.4947052154195
|
||||
31.62167800453515
|
||||
|
|
@ -0,0 +1,49 @@
|
|||
Song name : songs/Megalovania(240-7984)[filtered].wav
|
||||
Start : 7.984
|
||||
End : 14.984
|
||||
|
||||
Hit Objects :
|
||||
[8.078591836734693, 8.132503401360545]
|
||||
8.204578231292517
|
||||
[8.288115646258504, 8.34134693877551]
|
||||
8.527968253968254
|
||||
[8.879102040816326, 8.922888888888888]
|
||||
[9.129260770975057, 9.181766439909296]
|
||||
9.37977097505669
|
||||
9.423172335600906
|
||||
9.549498866213153
|
||||
9.689034013605442
|
||||
9.8049410430839
|
||||
9.898104308390023
|
||||
10.072764172335601
|
||||
10.276834467120182
|
||||
10.341426303854876
|
||||
10.512015873015873
|
||||
10.899748299319729
|
||||
11.129272108843537
|
||||
11.183319727891156
|
||||
[11.385530612244898, 11.428920634920635]
|
||||
11.57258276643991
|
||||
11.688943310657596
|
||||
11.76148299319728
|
||||
11.881256235827664
|
||||
11.935269841269841
|
||||
12.076947845804987
|
||||
12.293716553287982
|
||||
12.344793650793651
|
||||
[12.503886621315193, 12.549975056689341]
|
||||
[12.88151700680272, 12.925519274376416]
|
||||
13.12928344671202
|
||||
13.183308390022676
|
||||
13.420326530612245
|
||||
13.63055328798186
|
||||
13.695836734693877
|
||||
13.776233560090702
|
||||
13.816392290249432
|
||||
13.914839002267573
|
||||
14.015201814058956
|
||||
14.068489795918367
|
||||
14.142446712018142
|
||||
14.282956916099774
|
||||
14.527979591836734
|
||||
14.897560090702948
|
||||
|
|
@ -0,0 +1,97 @@
|
|||
Song name : audio.wav
|
||||
Start : 1.008
|
||||
End : 21.008
|
||||
|
||||
Hit Objects :
|
||||
[1.0518548752834467, 1.1318095238095238]
|
||||
1.4385102040816327
|
||||
1.8517641723356009
|
||||
1.9118095238095238
|
||||
2.1988276643990927
|
||||
2.590641723356009
|
||||
2.984496598639456
|
||||
3.3595192743764173
|
||||
3.403498866213152
|
||||
3.7352448979591837
|
||||
3.8218548752834467
|
||||
4.10879365079365
|
||||
4.192229024943311
|
||||
4.494666666666667
|
||||
4.894315192743765
|
||||
4.968136054421769
|
||||
5.245267573696145
|
||||
5.644451247165533
|
||||
5.823487528344671
|
||||
6.041390022675737
|
||||
6.4073197278911564
|
||||
6.785743764172335
|
||||
6.878181405895692
|
||||
6.985267573696145
|
||||
[7.173113378684807, 7.250256235827664]
|
||||
7.5618208616780045
|
||||
7.628351473922902
|
||||
7.929609977324263
|
||||
7.970482993197279
|
||||
8.309609977324264
|
||||
8.702648526077098
|
||||
9.100709750566892
|
||||
9.465505668934242
|
||||
9.65797732426304
|
||||
9.850721088435375
|
||||
10.227104308390022
|
||||
10.614780045351473
|
||||
10.673158730158729
|
||||
11.0098820861678
|
||||
11.083691609977325
|
||||
11.266696145124715
|
||||
11.379507936507938
|
||||
11.758986394557823
|
||||
12.156390022675737
|
||||
12.535868480725625
|
||||
12.714349206349205
|
||||
12.896265306122448
|
||||
13.279984126984125
|
||||
13.37145804988662
|
||||
13.665267573696145
|
||||
13.722784580498868
|
||||
14.048702947845804
|
||||
14.099213151927437
|
||||
14.140097505668933
|
||||
14.435789115646259
|
||||
14.817399092970522
|
||||
14.940732426303853
|
||||
15.216299319727892
|
||||
15.625063492063493
|
||||
15.759995464852608
|
||||
15.945392290249433
|
||||
16.147569160997733
|
||||
16.36139002267574
|
||||
16.444802721088436
|
||||
16.724621315192742
|
||||
17.101503401360542
|
||||
17.144802721088436
|
||||
17.187897959183672
|
||||
17.296344671201812
|
||||
17.379405895691608
|
||||
17.48303401360544
|
||||
17.875551020408164
|
||||
17.958532879818595
|
||||
18.27968934240363
|
||||
18.440879818594105
|
||||
18.653986394557823
|
||||
[18.829950113378683, 18.918034013605443]
|
||||
19.008544217687074
|
||||
19.084541950113376
|
||||
19.306492063492062
|
||||
19.388408163265304
|
||||
19.464836734693876
|
||||
19.77943990929705
|
||||
20.020301587301585
|
||||
20.167897959183673
|
||||
20.33250113378685
|
||||
20.402331065759636
|
||||
20.478589569160995
|
||||
20.53909977324263
|
||||
20.824961451247166
|
||||
20.869950113378685
|
||||
20.934825396825396
|
||||
|
|
@ -0,0 +1,61 @@
|
|||
Song name : audio.wav
|
||||
Start : 25.466
|
||||
End : 32.466
|
||||
|
||||
Hit Objects :
|
||||
25.52553514739229
|
||||
25.60707709750567
|
||||
25.738539682539685
|
||||
25.859526077097506
|
||||
25.942394557823132
|
||||
25.989775510204083
|
||||
26.09740589569161
|
||||
26.19691836734694
|
||||
26.363222222222223
|
||||
26.4916462585034
|
||||
26.554990929705216
|
||||
26.757621315192743
|
||||
26.866702947845805
|
||||
26.97207709750567
|
||||
27.072337868480727
|
||||
27.175580498866214
|
||||
27.399310657596374
|
||||
27.498664399092974
|
||||
27.716861678004538
|
||||
27.803505668934243
|
||||
27.902190476190476
|
||||
28.176839002267574
|
||||
28.273913832199547
|
||||
28.359049886621317
|
||||
28.46615873015873
|
||||
[28.562768707482995, 28.64804081632653]
|
||||
[28.756884353741498, 28.83251927437642]
|
||||
28.949106575963718
|
||||
29.01647619047619
|
||||
[29.084253968253968, 29.173789115646258]
|
||||
29.319333333333333
|
||||
29.41299546485261
|
||||
29.484344671201814
|
||||
29.559356009070296
|
||||
29.73783673469388
|
||||
29.926419501133786
|
||||
29.987961451247166
|
||||
30.09112471655329
|
||||
30.19709977324263
|
||||
30.301578231292517
|
||||
30.378426303854877
|
||||
30.481668934240364
|
||||
[30.683539682539685, 30.760727891156463]
|
||||
30.847587301587303
|
||||
31.04754195011338
|
||||
31.14048979591837
|
||||
31.229185941043085
|
||||
31.290421768707482
|
||||
31.433086167800454
|
||||
31.524730158730158
|
||||
31.61157823129252
|
||||
[31.72844897959184, 31.816873015873018]
|
||||
31.988358276643993
|
||||
32.095784580498865
|
||||
32.20156689342404
|
||||
32.29895918367347
|
||||
|
|
@ -0,0 +1,124 @@
|
|||
Song name : audio.wav
|
||||
Start : 35.659
|
||||
End : 55.659
|
||||
|
||||
Hit Objects :
|
||||
35.79228798185941
|
||||
35.91021315192744
|
||||
36.04037188208617
|
||||
36.15184580498866
|
||||
36.23495238095238
|
||||
36.32842176870748
|
||||
36.406290249433106
|
||||
36.75744671201814
|
||||
36.95995238095238
|
||||
37.2967664399093
|
||||
37.449056689342406
|
||||
37.5271746031746
|
||||
37.65802494331066
|
||||
37.84421541950113
|
||||
[37.947548752834464, 37.990655328798184]
|
||||
38.125190476190475
|
||||
38.411800453514736
|
||||
38.60766213151928
|
||||
38.70639229024943
|
||||
38.821278911564626
|
||||
38.8689433106576
|
||||
39.10621088435374
|
||||
39.28780952380952
|
||||
39.45389795918367
|
||||
39.684714285714286
|
||||
39.864
|
||||
39.920814058956914
|
||||
40.10293424036281
|
||||
40.195848072562356
|
||||
40.32245804988662
|
||||
40.56655102040816
|
||||
[40.74982766439909, 40.82168707482993]
|
||||
41.003024943310656
|
||||
41.14491836734694
|
||||
41.392820861678004
|
||||
41.57685714285714
|
||||
[41.72762811791383, 41.76821768707483]
|
||||
41.81646031746032
|
||||
41.87975963718821
|
||||
42.062174603174604
|
||||
42.31275283446712
|
||||
42.4768231292517
|
||||
42.53006575963719
|
||||
42.63855782312925
|
||||
42.81466893424036
|
||||
42.990655328798184
|
||||
43.133659863945574
|
||||
43.20110884353741
|
||||
43.258172335600904
|
||||
[43.56973696145125, 43.611335600907026]
|
||||
43.7077641723356
|
||||
44.05415873015873
|
||||
44.18567800453515
|
||||
44.2824126984127
|
||||
44.647401360544215
|
||||
44.76270748299319
|
||||
45.070519274376416
|
||||
45.406018140589566
|
||||
45.518750566893424
|
||||
45.694918367346936
|
||||
45.8018231292517
|
||||
46.045814058956914
|
||||
46.25060997732426
|
||||
46.47650566893424
|
||||
46.5686485260771
|
||||
[46.845791383219954, 46.91973696145125]
|
||||
47.06896598639456
|
||||
47.17215192743764
|
||||
47.29363718820861
|
||||
47.33420408163265
|
||||
47.43748072562358
|
||||
47.647129251700676
|
||||
47.71998639455782
|
||||
47.972854875283446
|
||||
48.054034013605445
|
||||
[48.258807256235826, 48.31286621315193]
|
||||
48.353263038548754
|
||||
48.48960090702948
|
||||
48.910995464852604
|
||||
[49.1158253968254, 49.16668707482993]
|
||||
49.26697052154195
|
||||
49.64987301587301
|
||||
49.74669841269841
|
||||
49.86911337868481
|
||||
50.1181723356009
|
||||
50.40005442176871
|
||||
50.459453514739224
|
||||
50.65728798185941
|
||||
50.71109750566893
|
||||
[50.84771882086168, 50.893149659863944]
|
||||
51.33664172335601
|
||||
51.39431746031746
|
||||
51.636709750566894
|
||||
51.688761904761904
|
||||
51.743988662131514
|
||||
52.00931746031746
|
||||
52.11760544217687
|
||||
52.15932879818594
|
||||
52.20068934240363
|
||||
52.25446485260771
|
||||
52.47726530612245
|
||||
52.55314965986395
|
||||
52.8780589569161
|
||||
53.04224263038549
|
||||
53.08240136054422
|
||||
53.177356009070294
|
||||
53.29026984126984
|
||||
53.49267346938775
|
||||
[53.62920408163265, 53.668081632653056]
|
||||
53.77542857142857
|
||||
[54.10584807256235, 54.16563265306122]
|
||||
54.327820861678006
|
||||
[54.431188208616774, 54.479566893424035]
|
||||
54.691052154195006
|
||||
54.992820861678
|
||||
55.19167573696145
|
||||
55.29097278911564
|
||||
55.553512471655324
|
||||
55.60739002267574
|
||||
145
script (1).py
145
script (1).py
|
|
@ -1,145 +0,0 @@
|
|||
import numpy as np
|
||||
import scipy as scp
|
||||
import heapq
|
||||
|
||||
def retrieve_dominant_freqs(song_name, offset, songlen, segsize):
|
||||
# returns a list with peak frequencies alongside the sample rate
|
||||
# /!\ song_name is specified to be a list, NOT a list of couples (aka song is mono)
|
||||
# segsize is in seconds
|
||||
|
||||
# remove high_pitched/low-pitched frequencies
|
||||
minfreq = 110
|
||||
maxfreq = 440*8
|
||||
|
||||
# cutting the song to only keep the one we're interested in
|
||||
subprocess.run(["ffmpeg", "-ss", str(offset), "-t", str(songlen+offset), "-i", song_name, "crop.wav"], shell=False)
|
||||
|
||||
# extracting data from cropped song
|
||||
sample_rate, song_data = wavfile.read("crop.wav")
|
||||
blit = int(sample_rate*segsize) # Te
|
||||
|
||||
# remove the copy of the song
|
||||
subprocess.run(["rm", "crop.wav"], shell=False)
|
||||
|
||||
# calculate the frequencies associated to the FFTs
|
||||
pfreq = scipy.fft.rfftfreq(blit, 1/sample_rate)
|
||||
|
||||
# left boundary of segment to crop
|
||||
current_time = offset
|
||||
|
||||
# list of FFTs
|
||||
fft_list = []
|
||||
|
||||
# number of samples
|
||||
k = 0
|
||||
|
||||
while(current_time <= songlen+offset):
|
||||
# index corresponding to left boundary
|
||||
left_id = int(current_time*sample_rate)
|
||||
|
||||
# index corresponding to right boundary
|
||||
right_id = int((current_time+segsize)*sample_rate)
|
||||
|
||||
# calculate the fft, append it to fft_list
|
||||
pff = scp.fft.rfft(global_data[left:right])
|
||||
fft_list.append(pff)
|
||||
|
||||
# just to avoid what causes 0.1 + 0.1 == 0.2 to be False
|
||||
k += 1
|
||||
current_time = offset + k*segsize
|
||||
|
||||
# spacing between samples (time)
|
||||
fe = segsize/sample_rate
|
||||
|
||||
# list that will contain the maximum frequencies/amplitudes for all FFTs
|
||||
maxlist = []
|
||||
maxamps = []
|
||||
|
||||
# find all maximums
|
||||
for i in range(len(fft_list)):
|
||||
current_max = -1
|
||||
current_fmax = 0
|
||||
|
||||
for j in range(len(fft_list[i])):
|
||||
if(pfreq[j] < maxfreq & pfreq[j] >= minfreq & np.abs(fft_list[i][j]) > current_max):
|
||||
current_max = np.abs(fft_list[i][j])
|
||||
current_fmax = pfreq[j]
|
||||
|
||||
maxlist.append(current_fmax)
|
||||
maxamps.append(current_max)
|
||||
|
||||
# gg
|
||||
# maxlist[i] corresponds to time (offset + i*segsize)
|
||||
return (maxlist, maxamps, segsize)
|
||||
|
||||
def retrieve_dominant_amps(song_name, offset, songlen, segsize, percent):
|
||||
# returns a list with the percent% peak amplitudes alongside the sample rate
|
||||
# /!\ song_name is specified to be a list, NOT a list of couples (aka song is mono)
|
||||
# segsize is in seconds
|
||||
|
||||
# cutting the song to only keep the one we're interested in
|
||||
subprocess.run(["ffmpeg", "-ss", str(offset), "-t", str(songlen+offset), "-i", song_name, "crop.wav"], shell=False)
|
||||
|
||||
# extracting data from cropped song
|
||||
sample_rate, song_data = wavfile.read("crop.wav")
|
||||
blit = int(sample_rate*segsize) # Te
|
||||
|
||||
# remove the copy of the song
|
||||
subprocess.run(["rm", "crop.wav"], shell=False)
|
||||
|
||||
# which notes will be voided
|
||||
is_locked = [False for i in range(len(song_data))]
|
||||
x = int((len(song_data)*threshold)//100)
|
||||
|
||||
print("Retreiving the", int(x), "/", len(song_data), "highest values")
|
||||
elements = heapq.nlargest(int(x), enumerate(song_data), key=lambda x: x[1])
|
||||
#returns a list of couples [id, value]
|
||||
|
||||
for idx in range(len(elements)):
|
||||
is_locked[elements[idx][0]] = True
|
||||
|
||||
for r in range(len(song_data)):
|
||||
if(is_locked[r] == False):
|
||||
song_data[r] = 0
|
||||
|
||||
# now we need to reduce song_data so that it matches the length of the previous function's return
|
||||
res = []
|
||||
k = 0
|
||||
current_time = offset
|
||||
|
||||
while(current_time <= songlen+offset):
|
||||
# index corresponding to left boundary
|
||||
left_id = int(current_time*sample_rate)
|
||||
|
||||
# index corresponding to right boundary
|
||||
right_id = int((current_time+segsize)*sample_rate)
|
||||
|
||||
# merge the segment into one value
|
||||
cmax = 0
|
||||
for i in range(left_id, right_id):
|
||||
if(i < len(song_data) & cmax < song_data[i]):
|
||||
cmax = song_data[i]
|
||||
|
||||
res.append(cmax)
|
||||
|
||||
k += 1
|
||||
current_time = current_time + k*segsize
|
||||
|
||||
# gg
|
||||
# res[i] corresponds to time (offset + i*segsize)
|
||||
return res
|
||||
|
||||
print("done")
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
872
sound_process.py
872
sound_process.py
|
|
@ -1,872 +0,0 @@
|
|||
from math import *
|
||||
import numpy as np
|
||||
from scipy.io import wavfile
|
||||
from scipy import signal
|
||||
import matplotlib.pyplot as plt
|
||||
import subprocess
|
||||
import wave as wv
|
||||
import struct
|
||||
import librosa
|
||||
import heapq
|
||||
import scipy
|
||||
import os
|
||||
import random
|
||||
from pathlib import Path
|
||||
from time import sleep
|
||||
from datetime import timedelta
|
||||
|
||||
import debug
|
||||
|
||||
print("Starting...\n")
|
||||
|
||||
def filter_n_percent_serial(song_name, offset, n_iter, step, threshold):
|
||||
"""
|
||||
song_name : string
|
||||
offset : int
|
||||
n_iter : int (number of turns)
|
||||
step : int (length of each small segment)
|
||||
threshold : int (is in ]0, 100])
|
||||
|
||||
filter data associated with song_name to keep only the highest threshold% values
|
||||
"""
|
||||
|
||||
subprocess.run(["ffmpeg", "-ss", str(offset), "-t", str(offset+step*n_iter), "-i", song_name, "crop.wav"], shell=False)
|
||||
|
||||
sample_rate, global_data = wavfile.read('crop.wav')
|
||||
|
||||
subprocess.run(["clear"], shell=False)
|
||||
subprocess.run(["rm", "crop.wav"], shell=False)
|
||||
|
||||
for i in range(n_iter):
|
||||
print(i, "/", n_iter)
|
||||
#print(i * step)
|
||||
song_data = global_data[int(i*step*sample_rate):int((i+1)*step*sample_rate)]
|
||||
|
||||
if(len(song_data) != 0):
|
||||
mx = max(song_data)
|
||||
|
||||
is_locked = [False for i in range(len(song_data))]
|
||||
x = int((len(song_data)*threshold)//100)
|
||||
#print("X = ", x)
|
||||
|
||||
#print("Retreiving the", int(x), "/", len(song_data), "highest values")
|
||||
elements = heapq.nlargest(int(x), enumerate(song_data), key=lambda x: x[1])
|
||||
#print("Done")
|
||||
|
||||
for idx in range(len(elements)):
|
||||
is_locked[elements[idx][0]] = True
|
||||
|
||||
for r in range(len(song_data)):
|
||||
if(is_locked[r] == False):
|
||||
global_data[r+int(i*step*sample_rate)] = 0
|
||||
|
||||
return global_data
|
||||
|
||||
|
||||
def write_to_file_thr(sample_rate, song_data, offset, threshold, filename):
|
||||
# write data to output file
|
||||
file = open(filename, 'w')
|
||||
file.writelines('time,amplitude\n')
|
||||
mx = max(song_data)
|
||||
print("writing to output...")
|
||||
for i in range(len(song_data)):
|
||||
if(i%(len(song_data)//50) == 0):
|
||||
print(i, "/", len(song_data))
|
||||
if(song_data[i]/mx > threshold):
|
||||
file.writelines(str(np.round(offset + i/sample_rate, 3)))
|
||||
file.writelines(',')
|
||||
file.writelines(str(np.round(song_data[i], 0)))
|
||||
file.writelines('\n')
|
||||
|
||||
def round_t(id, sample_rate, bpm, div, offset, k0):
|
||||
k = k0
|
||||
t = offset + k/(bpm*div)
|
||||
while(t < id/sample_rate):
|
||||
t = offset + k/(bpm*div)
|
||||
k += 1
|
||||
|
||||
if(np.abs(t - id/sample_rate) < np.abs((t - 1/(bpm*div)) - id/sample_rate)):
|
||||
return t
|
||||
return (t - 1/(bpm*div), 0)
|
||||
|
||||
def compress(Zxx):
|
||||
res = []
|
||||
|
||||
def get_freq(song_name, times, width=1000, display=False):
|
||||
"""
|
||||
for a given list of times (in seconds), returns the corresponding peak frequencies
|
||||
"""
|
||||
|
||||
subprocess.run(["ffmpeg", "-ss", str(0), "-t", str(max(np.array(times))), "-i", song_name, "crop.wav"], shell=False)
|
||||
|
||||
sample_rate, global_data = wavfile.read(song_name)
|
||||
#blit = int(sample_rate*step)
|
||||
|
||||
subprocess.run(["clear"], shell=False)
|
||||
subprocess.run(["rm", "crop.wav"], shell=False)
|
||||
|
||||
pfreq = scipy.fft.rfftfreq(2*width, 1/sample_rate)
|
||||
|
||||
frequencies = [0 for s in range(len(times))]
|
||||
print(len(pfreq))
|
||||
|
||||
for s in range(len(times)):
|
||||
left = max(0, int(times[s]*44100)-width)
|
||||
right = min(len(global_data), int(times[s]*44100)+width)
|
||||
pff = scipy.fft.rfft(global_data[left:right])
|
||||
|
||||
#print(len(pff), len(pfreq))
|
||||
|
||||
mx = max(np.abs(pff))
|
||||
for id in range(len(pff)):
|
||||
if frequencies[s] == 0 and np.abs(pff[id]) == mx:
|
||||
frequencies[s] = pfreq[id]
|
||||
|
||||
if(display):
|
||||
plt.plot(times, frequencies)
|
||||
plt.grid()
|
||||
plt.xlabel("Time (s)")
|
||||
plt.ylabel("Dominant frequency (Hz)")
|
||||
plt.title("Dominant frequencies at peaks")
|
||||
plt.show()
|
||||
|
||||
return frequencies
|
||||
|
||||
def is_data_stereo(raw_global_data:list) -> bool:
|
||||
"""
|
||||
raw_global_data : list
|
||||
"""
|
||||
try:
|
||||
assert(raw_global_data[0][0])
|
||||
except IndexError:
|
||||
return False
|
||||
except AssertionError:
|
||||
return True
|
||||
return True
|
||||
|
||||
|
||||
def void_freq(song_name, offset, songlen, increment, minfreq, maxfreq, upperthr, ampthr, ampfreq, ampval, leniency, write, linear, output_file="trimmed.wav"):
|
||||
"""
|
||||
song_name : string
|
||||
offset : int
|
||||
songlen : int (length of the part that will be filtered, starting from offset)
|
||||
increment : float (technical parameter)
|
||||
minfreq and maxfreq : every frequency in [minfreq, maxfreq] will be voided
|
||||
upperthr : every frequency above upperthr will be voided
|
||||
ampthr : every frequency with amplitude under MAX/ampthr (aka amplitudes under (100/ampthr)% of the max will be voided
|
||||
ampfreq, leniency (if linear is false), linear : technical parameters
|
||||
ampval : int
|
||||
- if linear is false, then this willbe the maximum amplification possible
|
||||
- if linear is true, this is the multiplier (Amp <- Amp * (ampval * frequency + leniency))
|
||||
write : bool (should be set to True)
|
||||
output_file : technical
|
||||
"""
|
||||
fft_list = []
|
||||
times = []
|
||||
current_time = offset
|
||||
k = 0
|
||||
|
||||
subprocess.run(["ffmpeg", "-ss", str(offset), "-t", str(songlen+offset), "-i", song_name, "crop.wav"], shell=False)
|
||||
|
||||
sample_rate, raw_global_data = wavfile.read("crop.wav")
|
||||
blit = int(sample_rate*increment)
|
||||
|
||||
global_data = [0 for i in range(len(raw_global_data))]
|
||||
|
||||
#subprocess.run(["clear"])
|
||||
subprocess.run(["rm", "crop.wav"], shell=False)
|
||||
|
||||
a = 0
|
||||
|
||||
if(is_data_stereo(raw_global_data)):
|
||||
print("Converting to mono...")
|
||||
for x in range(len(raw_global_data)):
|
||||
global_data[x] = raw_global_data[x][0]/2 + raw_global_data[x][1]/2
|
||||
|
||||
if(x % (int(len(raw_global_data)/100)) == 0):
|
||||
print(a, "/ 100")
|
||||
a += 1
|
||||
|
||||
else:
|
||||
global_data = raw_global_data
|
||||
|
||||
#print("Blit :", blit)
|
||||
|
||||
pfreq = scipy.fft.rfftfreq(blit, 1/sample_rate)
|
||||
|
||||
#print(len(pfreq))
|
||||
|
||||
while(current_time <= songlen):
|
||||
pff = scipy.fft.rfft(global_data[k*blit:(k+1)*blit])
|
||||
fft_list.append(pff)
|
||||
times.append(k*increment)
|
||||
|
||||
k += 1
|
||||
current_time = offset + k*increment
|
||||
|
||||
print("FFT :", len(fft_list), "\nFFT[0] :", len(fft_list[0]), "\npfreq :", len(pfreq))
|
||||
|
||||
|
||||
print("Finding global max...")
|
||||
|
||||
if(linear == False):
|
||||
for i in range(len(fft_list)):
|
||||
for j in range(len(fft_list[i])):
|
||||
fft_list[i][j] *= (1 + ampval/max(1, np.abs(pfreq[j] - ampfreq)))
|
||||
else:
|
||||
for i in range(len(fft_list)):
|
||||
for j in range(len(fft_list[i])):
|
||||
fft_list[i][j] *= (ampval*pfreq[j] + leniency)
|
||||
|
||||
print("Trimming...")
|
||||
|
||||
for i in range(len(fft_list)):
|
||||
lmax = 0
|
||||
for j in range(len(fft_list[i])):
|
||||
if(np.abs(fft_list[i][j]) > lmax):
|
||||
lmax = np.abs(fft_list[i][j])
|
||||
|
||||
for j in range(len(fft_list[i])):
|
||||
if((pfreq[j] >= minfreq and pfreq[j] < maxfreq) or pfreq[j] > upperthr):
|
||||
fft_list[i][j] = 0+0j
|
||||
|
||||
if(np.abs(fft_list[i][j]) < lmax/ampthr):
|
||||
fft_list[i][j] = 0+0j
|
||||
|
||||
|
||||
if(write):
|
||||
res = []
|
||||
print("Converting...")
|
||||
for i in range(len(fft_list)):
|
||||
ift = scipy.fft.irfft(fft_list[i], n=blit)
|
||||
for k in ift:
|
||||
res.append(k)
|
||||
#print(type(res[0]))
|
||||
mx = 0
|
||||
for j in range(len(res)):
|
||||
if(res[j] > mx):
|
||||
mx = res[j]
|
||||
|
||||
for i in range(len(res)):
|
||||
res[i] = np.int16(32767*res[i]/mx)
|
||||
|
||||
res = np.array(res)
|
||||
wavfile.write(output_file, 44100, res)
|
||||
|
||||
#plt.plot(np.abs(pfreq[:len(fft_list[0])]), np.abs(fft_list[0]))
|
||||
#plt.grid()
|
||||
#plt.show()
|
||||
|
||||
print("Done")
|
||||
|
||||
def convert_tuple(data, times):
|
||||
"""
|
||||
Takes data and converts it to a list of tuples (amplitude, datetimes)
|
||||
"""
|
||||
return [(times[i], data[i]) for i in range(len(data))]
|
||||
|
||||
def get_songlen(filename):
|
||||
"""
|
||||
retrieves the length of the song in seconds
|
||||
"""
|
||||
sample_rate, global_data = wavfile.read(filename)
|
||||
print("LEN :", len(global_data)/sample_rate)
|
||||
|
||||
return (len(global_data)/sample_rate)
|
||||
|
||||
def convert_to_wav(song_name:str, output_file="audio.wav") -> str:
|
||||
"""
|
||||
Converts the song to .wav, only if it's not already in wave format.
|
||||
Currently relies on file extension.
|
||||
Returns: the song_name that should be used afterwards.
|
||||
"""
|
||||
extension = Path(song_name).suffix
|
||||
match extension:
|
||||
case ".mp3" | ".ogg":
|
||||
print("Converting to .wav...")
|
||||
subprocess.run(["ffmpeg", "-y", "-i", song_name, output_file], shell=False)
|
||||
return output_file
|
||||
return song_name
|
||||
|
||||
def process_song(filename, bpm, offset0=0, div_len_factor=1, n_iter_2=-1, threshold=0.5, divisor=4):
|
||||
"""
|
||||
filename : string (name of the song)
|
||||
offset : int [+] (song mapping will start from this time in seconds, default is 0)
|
||||
bpm : int [+]
|
||||
div_len_factor : float [+] (the length multiplier of each segment, default is 1)
|
||||
n_iter : int [+*] (the number of iterations, default is -1 (maps the whole music))
|
||||
threshold : int [0, 100] (used by the filter function to only keep the largest threshold% of timing points, default is 0.5)
|
||||
divisor : int [+] (beat divisor used to snap the notes, default is 4)
|
||||
"""
|
||||
|
||||
filename = convert_to_wav(filename)
|
||||
|
||||
offset = offset0/1000
|
||||
|
||||
div_len = div_len_factor*60/bpm-0.01
|
||||
|
||||
n_iter = n_iter_2
|
||||
song_len = get_songlen(filename)
|
||||
|
||||
if(n_iter == -1):
|
||||
n_iter = int((song_len-offset/1000)/div_len)-1
|
||||
|
||||
filtered_name = f"{filename}_trimmed.wav"
|
||||
|
||||
void_freq(filename, offset, min(song_len, offset+div_len*(n_iter+1)+0.01), 4*60/bpm, minfreq=0, maxfreq=220, upperthr=5000, ampthr=60, ampfreq = 1200, ampval = 5.0, leniency = 0.005, write=True, linear=False, output_file=filtered_name)
|
||||
|
||||
datares = filter_n_percent_serial(filtered_name, offset, n_iter, div_len, threshold)
|
||||
|
||||
#snapped_data = amplitude
|
||||
#times in ms
|
||||
(snapped_data, times) = debug.snap3(datares, mintime=50, initial_plot=True, after_plot=True)
|
||||
|
||||
#frequencies=get_freq(filtered_name, offset, div_len, div_len*n_iter, snapped_data, True)
|
||||
frequencies = get_freq(filtered_name, times, display=True)
|
||||
|
||||
Path(f"{filename}_trimmed.wav").unlink()
|
||||
return snapped_data, times, frequencies
|
||||
|
||||
'''
|
||||
datares = debug.snap2(datares, 44100, bpm, first_offset=offset, div=divisor, show=True, adjust=True)
|
||||
frequencies = get_freq(filtered_name, offset, div_len, div_len*n_iter, datares, True)
|
||||
Path(f"{filename}_trimmed.wav").unlink()
|
||||
return convert_tuple(datares, frequencies)
|
||||
'''
|
||||
|
||||
def main():
|
||||
aa, bb, cc = process_song("tetris_4.wav", 160, n_iter_2=48)
|
||||
#print(data)
|
||||
print("Program finished with return 0")
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
''' -------------------------------------------------------------------- '''
|
||||
''' -----------------------| Feuilles mortes |-------------------------- '''
|
||||
''' -------------------------------------------------------------------- '''
|
||||
|
||||
|
||||
'''
|
||||
def smooth(data, thr, mergeThr, show):
|
||||
mx = max(data)
|
||||
for i in range(len(data)-mergeThr):
|
||||
if(data[i]/mx > thr):
|
||||
for k in range(1, mergeThr):
|
||||
data[i+k] = 0
|
||||
if(show):
|
||||
t = [j/1000 for j in range(len(data))]
|
||||
plt.plot(t, data)
|
||||
plt.xlabel("Time (not scaled to origin)")
|
||||
plt.ylabel("Amplitude")
|
||||
plt.grid()
|
||||
plt.show()
|
||||
|
||||
return data
|
||||
if(False):
|
||||
#t, f, Zxx = fct("no.wav", 0, 0.032, 10, 5000, False)
|
||||
#t, f, Zxx = fct("worlds_end_3.wav", 150.889, 0.032, 170.889, 3000, False)
|
||||
#t, f, Zxx = fct("deltamax.wav", 9.992, 0.032, 114.318, 3000, False)
|
||||
#t, f, Zxx = fct("deltamax.wav", 9.992, 0.032, 20, 3000, False)
|
||||
#t, f, Zxx = fct("da^9.wav", 8.463, 0.032, 20, 5000, False)
|
||||
t, f, Zxx = fct("13. Cosmic Mind.wav", 0, 0.032, 20, 5000, False)
|
||||
#t, f, Zxx = fct("Furioso Melodia 44100.wav", 4, 0.032, 8, 3000, False)
|
||||
#t, f, Zxx = fct("changing.wav", 0, 0.05, 3.9, 5000, False)
|
||||
#fct("worlds_end_3.wav", 75, (60/178)/4, 75+2, 2500)
|
||||
|
||||
plot_max(t, f, Zxx, True)
|
||||
|
||||
if(False):
|
||||
#(t, data) = peaks("worlds_end_3.wav", 0, 300, False, 0.92)
|
||||
(t, data) = peaks("worlds_end_3.wav", 74.582, 6, False, 0.9)
|
||||
#(t, data) = peaks("da^9.wav", 8.463, 301.924 - 8.463, False, 0.95)
|
||||
#(t, data) = peaks("deltamax.wav", 8.463, 30101.924 - 8.463, False, 0.92)
|
||||
da = find_bpm(t, 44100, data, 100, 200, 1, 10)
|
||||
print("BPM data is", da)'''
|
||||
|
||||
#data = [-1 for i in range(int(x))]
|
||||
#ids = [-1 for i in range(int(x))]
|
||||
'''
|
||||
data = []
|
||||
ids = []
|
||||
for k in range(int(x)):
|
||||
data.append(int(7*mx/10))
|
||||
ids.append(-1)
|
||||
# structure there is [[index, value]...]
|
||||
|
||||
i = 0
|
||||
calc = 0
|
||||
while(i < len(song_data)):
|
||||
if(i%10 == 0):
|
||||
print(i, "/", len(song_data))
|
||||
if(data[int(x)-1] < song_data[i]):
|
||||
calc += 1
|
||||
#print("\n \n \n \n \n")
|
||||
data[int(x)-1] = song_data[i]
|
||||
ids[int(x)-1] = i
|
||||
|
||||
k = int(x)-1
|
||||
#while(k < int(x) & data[0] > data[k]):
|
||||
while(k > 0 and data[k-1] <= data[k]):
|
||||
data[k], data[k-1] = data[k-1], data[k]
|
||||
ids[k], ids[k-1] = ids[k-1], ids[k]
|
||||
k -= 1
|
||||
|
||||
#print(data[int(x)-1], calc, "/", x)
|
||||
|
||||
i += skip
|
||||
i += 1
|
||||
|
||||
|
||||
for s in range(int(x)-1):
|
||||
if(data[s] < data[s+1]):
|
||||
print("Nope", s)
|
||||
assert(0)
|
||||
'''
|
||||
|
||||
|
||||
'''
|
||||
def fct(song_name, offset, increment, songlen, maxfreq, display):
|
||||
to_cut = 20000//maxfreq
|
||||
global_Zxx = np.array([])
|
||||
global_f = np.array([])
|
||||
global_t = np.array([])
|
||||
current_time = offset
|
||||
k = 0
|
||||
while(current_time <= songlen):
|
||||
subprocess.run(["ffmpeg", "-ss", str(current_time), "-t", str(increment), "-i", song_name, "crop.wav"], shell=False)
|
||||
|
||||
sample_rate, audio_data = wavfile.read('crop.wav')
|
||||
size = audio_data.size
|
||||
|
||||
#subprocess.run(["clear"])
|
||||
subprocess.run(["rm", "crop.wav"], shell=False)
|
||||
|
||||
# do stuff here
|
||||
#f, t, Zxx = signal.stft(audio_data, sample_rate, nperseg=1000)
|
||||
f, t, Zxx = signal.spectrogram(audio_data, fs=sample_rate, nfft=size)
|
||||
leng = len(f)
|
||||
|
||||
f, Zxx = f[:leng//to_cut], Zxx[:leng//to_cut]
|
||||
|
||||
#print(len(Zxx))
|
||||
#print(len(Zxx[0]))
|
||||
|
||||
|
||||
for i in range(len(Zxx)):
|
||||
for j in range(len(Zxx[i])):
|
||||
Zxx[i][j] *= 1127*np.log(1+f[i]/700)
|
||||
|
||||
|
||||
t = np.array([current_time + x for x in t])
|
||||
|
||||
if(k == 0):
|
||||
global_f = f
|
||||
global_t = t
|
||||
global_Zxx = Zxx
|
||||
else:
|
||||
global_Zxx = np.concatenate((global_Zxx, Zxx), axis=1)
|
||||
global_t = np.concatenate((global_t, t))
|
||||
|
||||
#print(len(global_t))
|
||||
|
||||
k += 1
|
||||
current_time = offset + k*increment
|
||||
|
||||
print("Completion rate : ", np.round(100*(current_time-offset)/(songlen-offset), 4), "%")
|
||||
|
||||
if(display):
|
||||
plt.pcolormesh(global_t, global_f, np.abs(global_Zxx), shading='gouraud')
|
||||
# print(len(global_Zxx), len(global_Zxx[0]))
|
||||
# 88 192 = 2500
|
||||
# 70 192 = 2000
|
||||
plt.title('STFT Magnitude')
|
||||
plt.ylabel('Frequency [Hz]')
|
||||
plt.xlabel('Time [sec]')
|
||||
plt.show()
|
||||
|
||||
return global_t, global_f, np.abs(global_Zxx)
|
||||
|
||||
def write_to_file(t, flist, maxlist, filename):
|
||||
file = open(filename, 'w')
|
||||
file.writelines('time,frequency,maxvalue\n')
|
||||
for i in range(len(t)):
|
||||
file.writelines(str(np.round(t[i], 3)))
|
||||
file.writelines(',')
|
||||
file.writelines(str(np.round(flist[i], 1)))
|
||||
file.writelines(',')
|
||||
file.writelines(str(np.round(maxlist[i], 0)))
|
||||
file.writelines('\n')
|
||||
#close(file)
|
||||
|
||||
def plot_max(time, freq, Zxx, save):
|
||||
fres = [0 for x in range(len(time))]
|
||||
maxres = [0 for x in range(len(time))]
|
||||
for t in range(len(time)):
|
||||
#subprocess.run(["clear"])
|
||||
print(t, "/", len(time))
|
||||
for f in range(len(Zxx)):
|
||||
if(maxres[t] < Zxx[f][t]):
|
||||
maxres[t] = Zxx[f][t]
|
||||
fres[t] = freq[f]
|
||||
|
||||
if(save):
|
||||
write_to_file(time, fres, maxres, 'output.csv')
|
||||
|
||||
''''''
|
||||
plt.plot(time, fres, 'r')
|
||||
plt.grid()
|
||||
plt.xlabel("Time")
|
||||
plt.ylabel("Maximum frequencies")
|
||||
|
||||
plt.plot(time, maxres, 'g')
|
||||
plt.grid()
|
||||
plt.xlabel("Time")
|
||||
plt.ylabel("Maximun values")
|
||||
|
||||
plt.show()''''''
|
||||
|
||||
fig, (ax1, ax2) = plt.subplots(2)
|
||||
fig.suptitle('Top : time and frequencies\nBottom : time and max values')
|
||||
ax1.plot(time, fres)
|
||||
ax2.plot(time, maxres)
|
||||
|
||||
plt.show()
|
||||
|
||||
def extract_peaks(song_data, sample_rate, offset, display, threshold):
|
||||
mx = max(song_data)
|
||||
for i in range(len(song_data)):
|
||||
#subprocess.run(["clear"])
|
||||
print(i, "/", len(song_data))
|
||||
if(song_data[i]/mx < threshold):
|
||||
song_data[i] = 0
|
||||
t = [offset + i/sample_rate for i in range(len(song_data))]
|
||||
|
||||
if(display):
|
||||
plt.plot(t, song_data, 'b+')
|
||||
plt.grid()
|
||||
plt.xlabel("t")
|
||||
plt.ylabel("amp")
|
||||
plt.show()
|
||||
|
||||
return (t, song_data)
|
||||
|
||||
def get_local_max(song_data, center, width):
|
||||
mx = 0
|
||||
for o in range(-width, width+1):
|
||||
togo = min(len(song_data)-1, center+o)
|
||||
togo = max(0, togo)
|
||||
if(mx < song_data[togo]):
|
||||
mx = song_data[togo]
|
||||
return mx
|
||||
|
||||
def extract_peaks_v2(song_data, sample_rate, offset, display, threshold, seglen):
|
||||
mx = 0
|
||||
for i in range(len(song_data)):
|
||||
if (i%seglen == 0):
|
||||
print("----")
|
||||
mx = get_local_max(song_data, i+seglen//2, seglen//2)
|
||||
#subprocess.run(["clear"])
|
||||
print(i, "/", len(song_data))
|
||||
if(song_data[i]/mx < threshold):
|
||||
song_data[i] = 0
|
||||
|
||||
t = [offset + i/sample_rate for i in range(len(song_data))]
|
||||
|
||||
if(display):
|
||||
plt.plot(t, song_data, 'b+')
|
||||
plt.grid()
|
||||
plt.xlabel("t")
|
||||
plt.ylabel("amp")
|
||||
plt.show()
|
||||
|
||||
return (t, song_data)
|
||||
|
||||
def peaks(song_name, offset, length, display, thr):
|
||||
subprocess.run(["ffmpeg", "-ss", str(offset), "-t", str(length), "-i", song_name, "crop.wav"], shell=False)
|
||||
|
||||
sample_rate, audio_data = wavfile.read('crop.wav')
|
||||
|
||||
#subprocess.run(["clear"])
|
||||
subprocess.run(["rm", "crop.wav"], shell=False)
|
||||
|
||||
#return extract_peaks(audio_data, sample_rate, offset, display, thr)
|
||||
return extract_peaks_v2(audio_data, sample_rate, offset, display, thr, 44100*2)
|
||||
|
||||
def find_bpm(sample_rate, data, minbpm, maxbpm, step, width):
|
||||
optimal = minbpm
|
||||
optimal_acc = 0
|
||||
accuracy = 0
|
||||
|
||||
bpmlst = []
|
||||
scores = []
|
||||
|
||||
for beat in range(minbpm, maxbpm+step, step):
|
||||
loopturn = 0
|
||||
print("testing", beat)
|
||||
accuracy = 0
|
||||
current = 0
|
||||
|
||||
while(current+width < len(data)):
|
||||
loopturn += 1
|
||||
for o in range(-width, width+1):
|
||||
accuracy += data[current + o]
|
||||
#current = (loopturn*sample_rate)//beat
|
||||
current += (sample_rate)//beat
|
||||
|
||||
#accuracy = accuracy/loopturn
|
||||
|
||||
#accuracy *= (1+(maxbpm-beat)/minbpm)
|
||||
if optimal_acc < accuracy:
|
||||
optimal_acc = accuracy
|
||||
optimal = beat
|
||||
bpmlst.append(beat)
|
||||
scores.append(accuracy)
|
||||
|
||||
if(False):
|
||||
plt.plot(bpmlst, scores)
|
||||
plt.xlabel("BPM")
|
||||
plt.ylabel("Score")
|
||||
plt.grid()
|
||||
plt.show()
|
||||
|
||||
return (optimal, optimal_acc)
|
||||
'''
|
||||
|
||||
|
||||
|
||||
'''
|
||||
def void_freq(song_name, offset, songlen, increment, lthr, gthr):
|
||||
to_cut = 20000//2500
|
||||
global_Zxx = np.array([])
|
||||
global_f = np.array([])
|
||||
global_t = np.array([])
|
||||
current_time = offset
|
||||
k = 0
|
||||
sample_rate, global_data = wavfile.read(song_name)
|
||||
blit = int(sample_rate*increment)
|
||||
print("Blit :", blit)
|
||||
while(current_time <= songlen):
|
||||
#subprocess.run(["ffmpeg", "-ss", str(current_time), "-t", str(increment), "-i", song_name, "crop.wav"])
|
||||
|
||||
#sample_rate, audio_data = wavfile.read('crop.wav')
|
||||
audio_data = global_data[int(k*blit):int((k+1)*blit)]
|
||||
size = audio_data.size
|
||||
|
||||
#subprocess.run(["clear"])
|
||||
#subprocess.run(["rm", "crop.wav"])
|
||||
|
||||
# do stuff here
|
||||
#f, t, Zxx = signal.stft(audio_data, sample_rate, nperseg=1000)
|
||||
f, t, Zxx = signal.spectrogram(audio_data, fs=sample_rate, nfft=size)
|
||||
leng = len(f)
|
||||
|
||||
f, Zxx = f[:leng//to_cut], Zxx[:leng//to_cut]
|
||||
|
||||
for i in range(len(Zxx)):
|
||||
for j in range(len(Zxx[i])):
|
||||
#Zxx[i][j] *= 1127*np.log(1+f[i]/700)
|
||||
Zxx[i][j] *= 1000
|
||||
|
||||
t = np.array([current_time + x for x in t])
|
||||
|
||||
if(k == 0):
|
||||
global_f = f
|
||||
global_t = t
|
||||
global_Zxx = Zxx
|
||||
else:
|
||||
global_Zxx = np.concatenate((global_Zxx, Zxx), axis=1)
|
||||
global_t = np.concatenate((global_t, t))
|
||||
|
||||
#print(len(global_t))
|
||||
|
||||
k += 1
|
||||
current_time = offset + k*increment
|
||||
|
||||
print("Completion rate : ", np.round(100*(current_time-offset)/(songlen-offset), 4), "%")
|
||||
|
||||
print("Finding global max...")
|
||||
gmax = 0
|
||||
for i in range(len(global_Zxx)):
|
||||
for j in range(len(global_Zxx[i])):
|
||||
if(global_Zxx[i][j] > gmax):
|
||||
gmax = global_Zxx[i][j]
|
||||
|
||||
print("Trimming...")
|
||||
for j in range(len(global_Zxx[0])):
|
||||
lmax = 0
|
||||
for i in range(len(global_Zxx)):
|
||||
if(global_Zxx[i][j] > lmax):
|
||||
lmax = global_Zxx[i][j]
|
||||
|
||||
for i in range(len(global_Zxx)):
|
||||
val = global_Zxx[i][j]
|
||||
if(val/lmax <= lthr/100):
|
||||
global_Zxx[i][j] = 0
|
||||
elif(val/gmax <= gthr/100):
|
||||
global_Zxx[i][j] = 0
|
||||
|
||||
if(False):
|
||||
print("Plotting...")
|
||||
plt.pcolormesh(global_t, global_f, np.abs(global_Zxx), shading='gouraud')
|
||||
# print(len(global_Zxx), len(global_Zxx[0]))
|
||||
print("XLEN :", len(global_Zxx), "\nYLEN :", len(global_Zxx[0]))
|
||||
plt.title('STFT Magnitude')
|
||||
plt.ylabel('Frequency [Hz]')
|
||||
plt.xlabel('Time [sec]')
|
||||
plt.show()
|
||||
|
||||
if(True):
|
||||
print("Converting...")
|
||||
audio_signal = librosa.griffinlim(global_Zxx)
|
||||
#scipy.io.wavfile.write('trimmed.wav', sample_rate, np.array(audio_signal, dtype=np.int16))
|
||||
wavfile.write('test.wav', sample_rate, np.array(audio_signal, dtype=np.int16))
|
||||
|
||||
print("Done")
|
||||
|
||||
def find_bpm_2(sample_rate, data, threshold, maxbpm, show):
|
||||
mx = np.max(data)
|
||||
min_spacing = (60*sample_rate)/maxbpm
|
||||
k = 0
|
||||
while(k < len(data) and data[k]/mx < threshold):
|
||||
k += 1
|
||||
|
||||
k += 1
|
||||
spacing = []
|
||||
current = 1
|
||||
progress = 0
|
||||
|
||||
while(k < len(data)):
|
||||
if(k%(len(data)/100) == 0):
|
||||
print(progress, "%")
|
||||
progress += 1
|
||||
if(data[k]/mx >= threshold and current > min_spacing):
|
||||
spacing.append(current)
|
||||
current = 0
|
||||
else:
|
||||
current += 1
|
||||
k += 1
|
||||
|
||||
|
||||
for x in range(len(spacing)):
|
||||
spacing[x] = 60/(spacing[x]/sample_rate)
|
||||
|
||||
digits = [i for i in range(len(spacing))]
|
||||
if(show):
|
||||
plt.plot(digits, spacing)
|
||||
plt.xlabel("N")
|
||||
plt.ylabel("BPM")
|
||||
plt.grid()
|
||||
plt.show()
|
||||
|
||||
beat = np.mean(spacing)
|
||||
error = np.std(spacing)
|
||||
|
||||
return (np.round(beat, 3), np.round(error, 3))
|
||||
|
||||
def to_ms(song_data, sample_rate, offset):
|
||||
# converts audio data to have exactly 1 sample per millisecond (aka set sample_rate to 1000)
|
||||
new_data = []
|
||||
spacing = int(sample_rate * 0.001)
|
||||
mx = max(song_data)
|
||||
i = 0
|
||||
while(i < len(song_data)):
|
||||
avg = 0
|
||||
for k in range(spacing):
|
||||
if(i+spacing < len(song_data)):
|
||||
avg += song_data[i+spacing]
|
||||
avg = avg / spacing
|
||||
new_data.append(avg)
|
||||
i += spacing
|
||||
|
||||
if(False): # pls dont kill me thx
|
||||
t = [offset + j/1000 for j in range(len(new_data))]
|
||||
plt.plot(t, new_data)
|
||||
plt.xlabel("Time")
|
||||
plt.ylabel("Amplitude")
|
||||
plt.grid()
|
||||
plt.show()
|
||||
|
||||
return (new_data, len(new_data))
|
||||
|
||||
def filter_n_percent(song_name, offset, length, threshold, reduce, show):
|
||||
# threshold is in ]0, 100]
|
||||
# filter data associated with song_name to keep only the highest threshold% values
|
||||
|
||||
subprocess.run(["ffmpeg", "-ss", str(offset), "-t", str(length), "-i", song_name, "crop.wav"], shell=False)
|
||||
|
||||
sample_rate, song_data = wavfile.read('crop.wav')
|
||||
|
||||
subprocess.run(["clear"], shell=False)
|
||||
subprocess.run(["rm", "crop.wav"], shell=False)
|
||||
|
||||
if(reduce):
|
||||
(song_data,e) = to_ms(song_data, 44100, 1)
|
||||
sample_rate = 1000
|
||||
|
||||
mx = max(song_data)
|
||||
|
||||
is_locked = [False for i in range(len(song_data))]
|
||||
x = int((len(song_data)*threshold)//100)
|
||||
#print("X = ", x)
|
||||
|
||||
print("Retreiving the", int(x), "/", len(song_data), "highest values")
|
||||
elements = heapq.nlargest(int(x), enumerate(song_data), key=lambda x: x[1])
|
||||
print("Done")
|
||||
|
||||
for idx in range(len(elements)):
|
||||
is_locked[elements[idx][0]] = True
|
||||
|
||||
for r in range(len(song_data)):
|
||||
if(is_locked[r] == False):
|
||||
song_data[r] = 0
|
||||
|
||||
if(show):
|
||||
#print("EEEEE")
|
||||
t = [offset + j/sample_rate for j in range(len(song_data))]
|
||||
plt.plot(t, song_data)
|
||||
plt.xlabel("Time")
|
||||
plt.ylabel("Amplitude")
|
||||
plt.grid()
|
||||
plt.show()
|
||||
|
||||
return song_data
|
||||
|
||||
def get_tpts(data, sample_rate, thr):
|
||||
res = []
|
||||
for i in range(len(data)):
|
||||
if(data[i] > thr):
|
||||
res.append(i/sample_rate)
|
||||
|
||||
for i in res:
|
||||
print(i)
|
||||
return res
|
||||
|
||||
def test_sample(timelist):
|
||||
for i in range(1,len(timelist)):
|
||||
#os.system('play -n synth %s sin %s' % (0.05, 440))
|
||||
for k in range(random.randint(1, 10)):
|
||||
print("E", end="")
|
||||
print("F")
|
||||
sleep(timelist[i]-timelist[i-1])
|
||||
'''
|
||||
Loading…
Reference in New Issue