def featureExtraction(soundfiles):
# extractor = esst.LowLevelSpectralExtractor()
extractor = esst.Extractor(dynamics = False,
dynamicsFrameSize = 88200,
dynamicsHopSize = 44100,
highLevel = False,
lowLevel = True,
lowLevelFrameSize = 2048,
lowLevelHopSize = 1024,
midLevel = True,
namespace = "",
relativeIoi = False,
rhythm = False,
sampleRate = 44100,
tonalFrameSize = 4096,
tonalHopSize = 2048,
tuning = True)
#soundfiles = listdir(inputPath)
for file in soundfiles:
path1= '/Users/helena/Desktop/SMC/ASP/sms-tools/workspace/A10/code/downloaded/'
name=file[70:-4] + '_features.json'
outPath = path1 + 'features/' + name
print file
audioLoader = esst.MonoLoader(filename=file)
audio = audioLoader()
pool = essentia.Pool()
pool = extractor(audio)
aggPool = esst.PoolAggregator()(pool)
output = esst.YamlOutput(filename = outPath, format='json')
output(aggPool)
print (outPath + ' exported')
def extractDefaultFeatures(audio, outputDir):
# compute all features for all sounds
extractor = ess.Extractor(dynamics=True,
dynamicsFrameSize=88200,
dynamicsHopSize=44100,
highLevel=True,
lowLevel=True,
lowLevelFrameSize=2048,
lowLevelHopSize=1024,
midLevel=True,
namespace="",
relativeIoi=False,
rhythm=True,
sampleRate=44100,
tonalFrameSize=4096,
tonalHopSize=2048,
tuning=True)
pool = essentia.Pool()
pool = extractor(audio)
aggPool = ess.PoolAggregator()(pool)
if not path.exists(outputDir):
makedirs(outputDir)
ess.YamlOutput(filename=outputDir + "features.json",
format="json",
doubleCheck=True)(aggPool)
def featureExtraction(soundfiles):
#extractor = esst.LowLevelSpectralExtractor()
extractor = esst.Extractor(dynamics=True,
dynamicsFrameSize=88200,
dynamicsHopSize=44100,
highLevel=True,
lowLevel=True,
lowLevelFrameSize=2048,
lowLevelHopSize=1024,
midLevel=True,
namespace="",
relativeIoi=False,
rhythm=True,
sampleRate=44100,
tonalFrameSize=4096,
tonalHopSize=2048,
tuning=True)
#soundfiles = listdir(inputPath)
for file, outPath in soundfiles:
audioLoader = esst.MonoLoader(filename=file)
audio = audioLoader()
pool = essentia.Pool()
pool = extractor(audio)
aggPool = esst.PoolAggregator()(pool)
esst.YamlOutput(filename=outPath + 'features.json',
format='json')(aggPool)
print(file + ' exported')
def extract_features_all(path, track_name):
features, features_frames = es.MusicExtractor(
lowlevelStats=['mean', 'stdev', 'min', 'max'],
rhythmStats=['mean', 'stdev', 'min', 'max'],
tonalStats=['mean', 'stdev', 'min', 'max'],
)(f'{path}{track_name}')
save_file = f'../data/essentia/{track_name[:-4]}.json'
es.YamlOutput(filename=save_file, format="json")(features)
def extract_features_intervals(path, track_name):
intervals = list(range(1, 31))
for i in intervals:
start = i - 1
end = i
features, features_frames = es.MusicExtractor(
lowlevelStats=['mean', 'stdev', 'min', 'max'],
rhythmStats=['mean', 'stdev', 'min', 'max'],
tonalStats=['mean', 'stdev', 'min', 'max'],
startTime=start,
endTime=end)(f'{path}{track_name}')
save_file = f'../data/essentia_bow/{track_name[:-4]}_{i}.json'
es.YamlOutput(filename=save_file, format="json")(features)
def savePoolToJson(input_pool, file_path="test_save.json"):
"""
Salva input_pool no .json definido por file_name,
se o arquivo já existir, sobreescreve-o.
O default de file_name serve para criar um arquivo de teste
"""
try:
remove(file_path)
except FileNotFoundError:
pass
es_mode.YamlOutput(filename=file_path,\
format="json",\
writeVersion=False,\
doubleCheck=True)(input_pool)
'lowlevelStats': stats,
'rhythmMethod': 'degara',
'rhythmMinTempo': 40,
'rhythmMaxTempo': 208,
'rhythmStats': stats,
'tonalFrameSize': 4096,
'tonalHopSize': 2048,
'tonalWindowType': 'blackmanharris62',
'tonalSilentFrames': 'noise',
'tonalStats': stats
}
features, features_frames = es.FreesoundExtractor(
**profile)(audio_in)
file_json = output + ".json"
es.YamlOutput(filename=file_json, format='json')(features)
tab_header = []
tab_value = []
features_name = sorted(features.descriptorNames())
for feat in range(0, len(features_name)):
name_feature = features_name[feat]
# print(name)
kind_of_feature = name_feature.split(".")
# print(type)
# ALL FEATURES
# if(kind_of_feature[0] != 'metadata' and
# kind_of_feature[1] != 'silence_rate_60dB' and
# kind_of_feature[1] != 'sound_start_frame' and
import confirm_prompt #for confirming user action
import json_flattener #for flatenning jsons
# [NOTE] Please set the path to dataset here(this is the path to directory where the instrument folders {bas, cel ... vio} will be moved in)
path_to_dataset='../../datasets/good-sounds/instruments'
instruments = os.listdir(path_to_dataset)
instruments.sort()
for instrument in instruments:
print("[Instrument] : " + instrument)
files=sorted(os.listdir(os.path.join(path_to_dataset,instrument)))
print("Number of files : "+str(len(files)))
for file in files:
if(file.endswith('.wav')):
filename=os.path.join(path_to_dataset,instrument,file)
print("Analysing file : "+filename)
# Compute all features, aggregate only 'mean' and 'stdev' statistics for all low-level, rhythm and tonal frame features
features, features_frames = es.MusicExtractor(lowlevelSilentFrames='drop',
lowlevelFrameSize=2048,
lowlevelHopSize=1024,
lowlevelStats=['mean', 'stdev'])(filename)
features_frames=[]
es.YamlOutput(filename = filename.replace('.wav','.json'), format='json')(features)
features=[]
filename=[]
print("Feature Extraction Completed Successfully")
'**************************************************************************'
)
INFO('processing entire audio from ' + str(startTime) + 's to ' +
str(endTime) + 's')
INFO(
'**************************************************************************'
)
INFO('Process step 3: Mid Level')
computeMidLevel(input_file, pool, startTime, endTime)
INFO('Process step 4: High Level')
highlevel.compute(pool)
else:
INFO('Process step 2: Low Level')
computeLowLevel(input_file, pool, startTime, endTime)
INFO('Process step 3: Mid Level')
computeMidLevel(input_file, pool, startTime, endTime)
INFO('Process step 4: High Level')
highlevel.compute(pool)
# compute statistics
INFO('Process step 5: Aggregation')
stats = computeAggregation(pool, segments_namespace)
# svm
#addSVMDescriptors(stats, opt.svmpath)
# output results to file
INFO('writing results to ' + output_file)
standard.YamlOutput(filename=output_file)(stats)
INFO('Process step 4: High Level')
highlevel.compute(eqPool)
highlevel.compute(neqPool)
else:
INFO('Process step 2: Low Level')
computeLowLevel(input_file, neqPool, eqPool, startTime, endTime)
INFO('Process step 3: Mid Level')
computeMidLevel(input_file, neqPool, eqPool, startTime, endTime)
INFO('Process step 4: High Level')
highlevel.compute(eqPool)
highlevel.compute(neqPool)
# compute statistics
INFO('Process step 5: Aggregation')
neqStats = computeAggregation(neqPool, segments_namespace)
eqStats = computeAggregation(eqPool, segments_namespace)
# svm
#addSVMDescriptors(neqStats, opt.svmpath)
# output results to file
eqStats.set('metadata.audio_properties.equal_loudness', 1)
neqStats.set('metadata.audio_properties.equal_loudness', 0)
INFO('writing results to ' + output_file)
standard.YamlOutput(filename=output_file)(eqStats)
(name, ext) = os.path.splitext(output_file)
output_file = name + '..neq.sig'
INFO('writing results to ' + output_file)
standard.YamlOutput(filename=output_file)(neqStats)
"""
# So let's redo the previous using a Pool
pool = es.Pool()
for frame in ess.FrameGenerator(audio, frameSize=1024, hopSize=512):
mfcc_bands, mfcc_coeffs = mfcc(spectrum(w(frame)))
pool.add('lowlevel.mfcc', mfcc_coeffs)
pool.add('lowlevel.mfcc_bands', mfcc_bands)
"""
plotMfcc = pool['lowlevel.mfcc'].T[1:,:]
plt.pcolormesh(plotMfcc)
plt.show()
"""
#output = es.YamlOutput(filename = 'mfcc.sig')
output = ess.YamlOutput(filename='joeTestOut/mfcc.json', format='json')
output(pool)
# Say we're not interested in all the MFCC frames, but just their mean & variance.
# To this end, we have the PoolAggregator algorithm, that can do all sorts of
# aggregation: mean, variance, min, max, etc...
aggrPool = ess.PoolAggregator(defaultStats=['mean', 'var'])(pool)
print 'Original pool descriptor names:'
print pool.descriptorNames()
print
print 'Aggregated pool descriptor names:'
print aggrPool.descriptorNames()
output = ess.YamlOutput(filename='joeTestOut/mfccaggr.json', format='json')
output(aggrPool)
def main():
aparser = argparse.ArgumentParser()
aparser.add_argument(
'-c',
action='store',
dest='config',
help=
'-c type of the dataset. For ex: _1s_h100 for 1s with full length hop')
aparser.add_argument('-t',
action='store',
dest='data_type',
help='-t type of data original/harmonic/residual')
args = aparser.parse_args()
if not args.config:
aparser.error('Please specify the data config!')
conf = args.config
if args.data_type == 'original':
path_to_dataset = PATH_TO_ORIGINAL_WAV_FILES + conf
path_to_features = PATH_TO_ORIGINAL_FEATURES + conf
if args.data_type == 'residual':
path_to_dataset = PATH_TO_RESIDUAL_WAV_FILES + conf
path_to_features = PATH_TO_RESIDUAL_FEATURES + conf
if args.data_type == 'harmonic':
path_to_dataset = PATH_TO_HARMONIC_WAV_FILES + conf
path_to_features = PATH_TO_HARMONIC_FEATURES + conf
datasets = sorted(os.listdir(path_to_dataset))
for dataset in datasets:
empty_files = 0
print("[Dataset] : " + dataset)
folder_path = os.path.join(path_to_dataset, dataset)
lrms = sorted(os.listdir(folder_path))
for channel in lrms:
channel_path = os.path.join(folder_path, channel)
sub_folders = sorted(os.listdir(channel_path))
for sub_folder in sub_folders:
sub_folder_path = os.path.join(channel_path, sub_folder)
files = sorted(os.listdir(sub_folder_path))
for filename in files:
filepath = os.path.join(sub_folder_path, filename)
features = essentia.Pool()
try:
# Compute all features, aggregate only 'mean' and 'stdev' statistics for all low-level, rhythm and tonal frame features
features, features_frames = es.MusicExtractor(
lowlevelSilentFrames='drop',
lowlevelFrameSize=2048,
lowlevelHopSize=1024,
lowlevelStats=['mean', 'stdev'])(filepath)
features_frames = []
except RuntimeError, e:
print(filepath + " is almost silent")
empty_files += 1
dump_path = os.path.join(path_to_features, dataset,
channel, sub_folder)
create_folder(dump_path)
es.YamlOutput(filename=os.path.join(
dump_path, filename.replace('.wav', '.json')),
format='json')(features)
features = []
filename = []
print("Feature Extraction Completed Successfully for " + dataset)
print("Total number of empty file in " + dataset + " is " +
str(empty_files))
