def build():
filename = 'FlavioGaete22.zip'
# stream all the audio files from the zip archive
print 'Processing Audio...'
for zf in ff.iter_zip(filename):
if '._' in zf.filename:
continue
print zf.filename
WithTimbre.process(meta=zf)
# learn k-means codes for the bfcc frames
print 'Learning K-Means clusters...'
BfccKmeans.process(docs=(wt.bfcc for wt in WithTimbre))
# build an index
print 'Building index...'
BfccKmeansIndex.build()
def test_can_handle_zip_file(self):
STFT = stft(resample_to=SR11025(), store_fft=True)
@simple_in_memory_settings
class Document(STFT):
pass
signal = SineSynthesizer(SR11025()).synthesize(Seconds(2))
bio = BytesIO()
filename = 'test.wav'
with zipfile.ZipFile(bio, mode='w') as zf:
zf.writestr(filename, signal.encode().read())
bio.seek(0)
zip_wrapper = featureflow.iter_zip(bio).next()
_id = Document.process(meta=zip_wrapper)
doc = Document(_id)
self.assertEqual(2, doc.ogg.duration_seconds)
# forward and backward transformations
pipeline = ff.PickleFeature(
zounds.PreprocessingPipeline,
needs=(log, unit_norm, kmeans),
store=True)
if __name__ == '__main__':
# stream all the audio files from the zip archive
# you can download the original file here:
# https://archive.org/details/FlavioGaete
# - https://archive.org/download/FlavioGaete/FlavioGaete22.zip
filename = 'FlavioGaete22.zip'
print 'Processing Audio...'
for zf in ff.iter_zip(filename):
if '._' in zf.filename:
continue
try:
# check if the feature already exists
Document(zf.filename).mdct.shape
except KeyError:
print 'processing {filename}'.format(filename=zf.filename)
Document.process(meta=zf, _id=zf.filename)
print 'learn k-means clusters'
DctKmeans.process(docs=(doc.mdct for doc in Document))
print 'learn k-means clusters with log amplitude'
DctKmeansWithLogAmplitude.process(docs=(doc.mdct for doc in Document))
print('Downloading {url} -> {filename}...'.format(**locals()))
with open(filename, 'wb') as f:
for chunk in resp.iter_content(chunk_size=1000000):
f.write(chunk)
return filename
if __name__ == '__main__':
index = zounds.HammingIndex(
WithCodes, WithCodes.bfcc_kmeans_pooled, listen=True)
zip_filename = download_zip_archive()
print('Processing Audio...')
for zf in ff.iter_zip(zip_filename):
if '._' in zf.filename:
continue
try:
print('processing {zf.filename}'.format(**locals()))
WithTimbre.process(
_id=zf.filename, meta=zf, raise_if_exists=True)
except ff.ModelExistsError as e:
print(e)
# learn K-Means centroids
try:
print('learning K-Means centroids')
BfccKmeans.process(