forked from gdrtodd/CS236-Project
-
Notifications
You must be signed in to change notification settings - Fork 0
/
harmonicity_evaluation.py
214 lines (169 loc) · 8.08 KB
/
harmonicity_evaluation.py
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
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
"""
Evaluate the harmonicity between bass and piano tracks for different
datasets. Uses code adapted from pypianoroll.
"""
import glob, os
from pypianoroll import metrics, Multitrack
import numpy as np
import warnings
# <Pypianoroll code>
def get_tonal_matrix(r1=1.0, r2=1.0, r3=0.5):
"""Compute and return a tonal matrix for computing the tonal distance [1].
Default argument values are set as suggested by the paper.
[1] Christopher Harte, Mark Sandler, and Martin Gasser. Detecting harmonic
change in musical audio. In Proc. ACM MM Workshop on Audio and Music
Computing Multimedia, 2006.
"""
tonal_matrix = np.empty((6, 12))
tonal_matrix[0] = r1 * np.sin(np.arange(12) * (7. / 6.) * np.pi)
tonal_matrix[1] = r1 * np.cos(np.arange(12) * (7. / 6.) * np.pi)
tonal_matrix[2] = r2 * np.sin(np.arange(12) * (3. / 2.) * np.pi)
tonal_matrix[3] = r2 * np.cos(np.arange(12) * (3. / 2.) * np.pi)
tonal_matrix[4] = r3 * np.sin(np.arange(12) * (2. / 3.) * np.pi)
tonal_matrix[5] = r3 * np.cos(np.arange(12) * (2. / 3.) * np.pi)
return tonal_matrix
def get_num_pitch_used(pianoroll):
"""Return the number of unique pitches used in a piano-roll."""
return np.sum(np.sum(pianoroll, 0) > 0)
def get_qualified_note_rate(pianoroll, threshold=2):
"""Return the ratio of the number of the qualified notes (notes longer than
`threshold` (in time step)) to the total number of notes in a piano-roll."""
padded = np.pad(pianoroll.astype(int), ((1, 1), (0, 0)), 'constant')
diff = np.diff(padded, axis=0)
flattened = diff.T.reshape(-1,)
onsets = (flattened > 0).nonzero()[0]
offsets = (flattened < 0).nonzero()[0]
num_qualified_note = (offsets - onsets >= threshold).sum()
return num_qualified_note / len(onsets)
def get_polyphonic_ratio(pianoroll, threshold=2):
"""Return the ratio of the number of time steps where the number of pitches
being played is larger than `threshold` to the total number of time steps"""
return np.sum(np.sum(pianoroll, 1) >= threshold) / pianoroll.shape[0]
def get_in_scale(chroma, scale_mask=None):
"""Return the ratio of chroma."""
measure_chroma = np.sum(chroma, axis=0)
in_scale = np.sum(np.multiply(measure_chroma, scale_mask, dtype=float))
return in_scale / np.sum(chroma)
def get_drum_pattern(measure, drum_filter):
"""Return the drum_pattern metric value."""
padded = np.pad(measure, ((1, 0), (0, 0)), 'constant')
measure = np.diff(padded, axis=0)
measure[measure < 0] = 0
max_score = 0
for i in range(6):
cdf = np.roll(drum_filter, i)
score = np.sum(np.multiply(cdf, np.sum(measure, 1)))
if score > max_score:
max_score = score
return max_score / np.sum(measure)
def get_harmonicity(bar_chroma1, bar_chroma2, resolution, tonal_matrix=None):
"""Return the harmonicity metric value"""
if tonal_matrix is None:
tonal_matrix = get_tonal_matrix()
warnings.warn("`tonal matrix` not specified. Use default tonal matrix",
RuntimeWarning)
score_list = []
for r in range(bar_chroma1.shape[0]//resolution):
start = r * resolution
end = (r + 1) * resolution
beat_chroma1 = np.sum(bar_chroma1[start:end], 0)
beat_chroma2 = np.sum(bar_chroma2[start:end], 0)
score_list.append(tonal_dist(beat_chroma1, beat_chroma2, tonal_matrix))
score_list = np.array(score_list)
return np.mean(score_list[~np.isnan(score_list)])
def to_chroma(pianoroll):
"""Return the chroma features (not normalized)."""
padded = np.pad(pianoroll, ((0, 0), (0, 12 - pianoroll.shape[1] % 12)),
'constant')
return np.sum(np.reshape(padded, (pianoroll.shape[0], 12, -1)), 2)
def tonal_dist(chroma1, chroma2, tonal_matrix=None):
"""Return the tonal distance between two chroma features."""
if tonal_matrix is None:
tonal_matrix = get_tonal_matrix()
warnings.warn("`tonal matrix` not specified. Use default tonal matrix",
RuntimeWarning)
with warnings.catch_warnings():
warnings.simplefilter("ignore", category=RuntimeWarning)
chroma1 = chroma1 / np.sum(chroma1)
chroma2 = chroma2 / np.sum(chroma2)
result1 = np.matmul(tonal_matrix, chroma1)
result2 = np.matmul(tonal_matrix, chroma2)
return np.linalg.norm(result1 - result2)
# </Pypianoroll code>
def get_dataset_metrics(dir="./generated_samples", base_term="unconditional"):
"""
Get harmonicity metric for a dataset provided by a directory and base_term.
Assumes samples are of the form <base_term>_<index>_<bass/melody>.mid.
"""
bass_metrics = np.zeros((100, 3))
melody_metrics = np.zeros((100, 3))
harmonicity = np.zeros(100)
tonal_matrix = get_tonal_matrix()
for array_index, track_index in enumerate(range(0, 300, 3)):
bass_track = Multitrack(os.path.join(dir, "{}_{}_bass.mid".format(base_term, track_index)))
melody_track = Multitrack(os.path.join(dir, "{}_{}_melody.mid".format(base_term, track_index)))
bass_pianoroll = bass_track.tracks[0].pianoroll
melody_pianoroll = melody_track.tracks[0].pianoroll
chroma_bass = to_chroma(bass_pianoroll)
chroma_melody = to_chroma(melody_pianoroll)
harmonicity[array_index] = get_harmonicity(chroma_melody, chroma_bass, 50,
tonal_matrix = tonal_matrix)
bass_metrics[array_index,:] = get_metrics(bass_pianoroll)
melody_metrics[array_index, :] = get_metrics(melody_pianoroll)
return harmonicity
def get_ground_truth_metrics(dir="./data_processed/midis_tracks=Bass-Piano"):
"""
Get harmonicity metric for a ground-truth dataset. Assumes songs have distinct
bass and piano tracks that are recognizable by pypianoroll's Multitrack object.
"""
bass_metrics = np.zeros((100, 3))
melody_metrics = np.zeros((100, 3))
harmonicity = np.zeros(100)
tonal_matrix = get_tonal_matrix()
array_index = 0
for track in glob.glob(os.path.join(dir,"*.mid")):
try:
comb_track = Multitrack(track)
first_track_name = comb_track.tracks[0].name
second_track_name = comb_track.tracks[1].name
except:
continue
if first_track_name == "Piano" and second_track_name == "Bass":
melody_pianoroll = comb_track.tracks[0].pianoroll
bass_pianoroll = comb_track.tracks[1].pianoroll
elif first_track_name == "Bass" and second_track_name == "Piano":
melody_pianoroll = comb_track.tracks[1].pianoroll
bass_pianoroll = comb_track.tracks[0].pianoroll
else:
print(first_track_name, second_track_name)
continue
chroma_bass = to_chroma(bass_pianoroll)
chroma_melody = to_chroma(melody_pianoroll)
harmonicity[array_index] = get_harmonicity(chroma_melody, chroma_bass, 50,
tonal_matrix = tonal_matrix)
bass_metrics[array_index,:] = get_metrics(bass_pianoroll)
melody_metrics[array_index, :] = get_metrics(melody_pianoroll)
array_index += 1
if array_index >= 100:
break
return harmonicity[:array_index]
def get_metrics(pianoroll):
"""
Helper function to compute some of the other evaluation metrics such
as number of pitches used, number of pitch classes used, and the rate
of polyphonicity.
"""
n_pitches = metrics.n_pitches_used(pianoroll)
n_classes = metrics.n_pitche_classes_used(pianoroll)
poly_rate = metrics.polyphonic_rate(pianoroll, threshold=1)
return np.array([n_pitches, n_classes, poly_rate])
if __name__ == "__main__":
ground_truth_metrics = get_ground_truth_metrics()
unconditional_metrics = get_dataset_metrics(base_term="unconditional")
conditional_metrics = get_dataset_metrics(base_term="conditional")
# Plot tonal distance distributions
import matplotlib.pyplot as plt
plt.hist([unconditional_metrics, conditional_metrics, ground_truth_metrics])
plt.xlabel("Tonal Distance")
plt.ylabel("Count")
plt.show()