Closed Hats

Infilling HiHats (Symbolic Vs Audio Versions)

Below Are the examples for the Infilling HiHats Models comparing the symbolic model (IHS) vs the audio one (IH).

More examples can be found at https://github.com/AnonUserGit/AnonUserGit.github.io/tree/master/assets/wav

---
GT: Ground Truth (Complete Version from which HiHats were removed during training)
IH: Generated Hihats (Audio Input)
IHS: Generated Hihats (Symbolic Input)
IH (Mix): Generated HiHats Mixed with Inputs (Audio Input)
IHS: Generated HiHats Mixed with Inputs (Symbolic Input)
---

NOTE that the generated events are quieter than the ground truth. No adjustment was made for creating the mixed versions to demonstrate the performance of the model in replicating velocities.

INDEX GT IH IHS IH (Mix) IHS (Mix)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60