def get_dataflow(cls, params, samplerate):
df = Frames.get_dataflow(Frames.filter_params(params), samplerate)
if (params['FFTLength'] == '0'):
params['FFTLength'] = params['blockSize']
dataflow_safe_append(df, 'FFT', params)
dataflow_safe_append(df, 'Abs', {})
return df
def get_dataflow(cls, params, samplerate):
df = Frames.get_dataflow(Frames.filter_params(params), samplerate)
if params["FFTLength"] == "0":
params["FFTLength"] = params["blockSize"]
dataflow_safe_append(df, "FFT", params)
dataflow_safe_append(df, "Abs", {})
return df
def get_dataflow(cls, params, samplerate):
df = Frames.get_dataflow(Frames.filter_params(params), samplerate)
if (params['FFTLength'] == '0'):
params['FFTLength'] = params['blockSize']
dataflow_safe_append(df, 'FFT', params)
dataflow_safe_append(df, 'Abs', {})
return df
def get_dataflow(cls, params, samplerate):
nbCoeffs = int(params.get('LSFNbCoeffs'))
displacement = int(params.get('LSFDisplacement'))
lpcparams = LPC.filter_params(params)
lpcparams['LPCNbCoeffs'] = nbCoeffs + 1 - max(displacement, 1)
df = LPC.get_dataflow(lpcparams, samplerate)
dataflow_safe_append(df, 'LPC2LSF', params)
return df
def get_dataflow(cls, params, samplerate):
nbCoeffs = int(params.get('LSFNbCoeffs'))
displacement = int(params.get('LSFDisplacement'))
lpcparams = LPC.filter_params(params)
lpcparams['LPCNbCoeffs'] = nbCoeffs + 1 - max(displacement, 1)
df = LPC.get_dataflow(lpcparams, samplerate)
dataflow_safe_append(df, 'LPC2LSF', params)
return df
def get_dataflow(cls, params, samplerate):
minFreq = float(params["CQTMinFreq"])
bins = int(params["CQTBinsPerOctave"])
Q = 2.0 / (pow(2.0, 1.0 / bins) - 1)
fftLen = Q * samplerate / minFreq
fftLen = pow(2, math.ceil(math.log(fftLen) / math.log(2)))
fParams = Frames.filter_params(params)
fParams["blockSize"] = "%i" % fftLen
df = Frames.get_dataflow(fParams, samplerate)
dataflow_safe_append(df, "FFT", {"FFTLength": "%i" % fftLen, "FFTWindow": "None"})
dataflow_safe_append(df, "CQT", params)
return df
def get_dataflow(cls, params, samplerate):
minFreq = float(params['CQTMinFreq'])
bins = int(params['CQTBinsPerOctave'])
Q = 2.0 / (pow(2.0, 1.0 / bins) - 1)
fftLen = Q * samplerate / minFreq
fftLen = pow(2, math.ceil(math.log(fftLen) / math.log(2)))
fParams = Frames.filter_params(params)
fParams['blockSize'] = '%i' % fftLen
df = Frames.get_dataflow(fParams, samplerate)
dataflow_safe_append(
df, 'FFT', {'FFTLength': '%i' % fftLen, 'FFTWindow': 'None'})
dataflow_safe_append(df, 'CQT', params)
return df
def get_dataflow(cls, params, samplerate):
minFreq = float(params['CQTMinFreq'])
bins = int(params['CQTBinsPerOctave'])
Q = 2.0 / (pow(2.0, 1.0 / bins) - 1)
fftLen = Q * samplerate / minFreq
fftLen = pow(2, math.ceil(math.log(fftLen) / math.log(2)))
fParams = Frames.filter_params(params)
fParams['blockSize'] = '%i' % fftLen
df = Frames.get_dataflow(fParams, samplerate)
dataflow_safe_append(
df, 'FFT', {'FFTLength': '%i' % fftLen, 'FFTWindow': 'None'})
dataflow_safe_append(df, 'CQT', params)
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(MagnitudeSpectrum.filter_params(params), samplerate)
dataflow_safe_append(df, "Sqr", {})
dataflow_safe_append(df, "Loudness", params)
if params["LMode"] == "Relative":
dataflow_safe_append(df, "Normalize", {"NNorm": "Sum"})
elif params["LMode"] == "Total":
dataflow_safe_append(df, "Sum", {})
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(
MagnitudeSpectrum.filter_params(params), samplerate)
dataflow_safe_append(df, 'Sqr', {})
dataflow_safe_append(df, 'Loudness', params)
if (params['LMode'] == 'Relative'):
dataflow_safe_append(df, 'Normalize', {'NNorm': 'Sum'})
elif (params['LMode'] == 'Total'):
dataflow_safe_append(df, 'Sum', {})
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(
MagnitudeSpectrum.filter_params(params), samplerate)
dataflow_safe_append(df, 'Sqr', {})
dataflow_safe_append(df, 'Loudness', params)
if (params['LMode'] == 'Relative'):
dataflow_safe_append(df, 'Normalize', {'NNorm': 'Sum'})
elif (params['LMode'] == 'Total'):
dataflow_safe_append(df, 'Sum', {})
return df
def get_dataflow(cls, params, samplerate):
tuning = float(params['CZTuning'])
fmin = float(params['CQTMinFreq'])
if tuning > 0:
# adjust min freq to a divisor of tuning
b = int(params['CQTBinsPerOctave'])
if (b % 12 != 0):
print 'WARNING: in Chroma2, CQTBinsPerOctave must be multiple of 12'
b = b - b % 12
if b == 0:
b = 12
print 'use CQTBinsPerOctave=%i' % b
dev = b * math.log(tuning / fmin) / math.log(2)
fmin *= pow(2.0, math.fmod(dev, 1) / b)
params['CQTMinFreq'] = str(fmin)
df = CQT.get_dataflow(CQT.filter_params(params), samplerate)
chParams = params
del chParams['stepSize']
dataflow_safe_append(df, 'Chroma2', chParams)
return df
def get_dataflow(cls, params, samplerate):
tuning = float(params["CZTuning"])
fmin = float(params["CQTMinFreq"])
if tuning > 0:
# adjust min freq to a divisor of tuning
b = int(params["CQTBinsPerOctave"])
if b % 12 != 0:
print "WARNING: in Chroma2, CQTBinsPerOctave must be multiple of 12"
b = b - b % 12
if b == 0:
b = 12
print "use CQTBinsPerOctave=%i" % b
dev = b * math.log(tuning / fmin) / math.log(2)
fmin *= pow(2.0, math.fmod(dev, 1) / b)
params["CQTMinFreq"] = str(fmin)
df = CQT.get_dataflow(CQT.filter_params(params), samplerate)
chParams = params
del chParams["stepSize"]
dataflow_safe_append(df, "Chroma2", chParams)
return df
def get_dataflow(cls, params, samplerate):
tuning = float(params['CZTuning'])
fmin = float(params['CQTMinFreq'])
if tuning > 0:
# adjust min freq to a divisor of tuning
b = int(params['CQTBinsPerOctave'])
if (b % 12 != 0):
print('WARNING: in Chroma2, CQTBinsPerOctave must be multiple of 12')
b = b - b % 12
if b == 0:
b = 12
print('use CQTBinsPerOctave=%i' % b)
dev = b * math.log(tuning / fmin) / math.log(2)
fmin *= pow(2.0, math.fmod(dev, 1) / b)
params['CQTMinFreq'] = str(fmin)
df = CQT.get_dataflow(CQT.filter_params(params), samplerate)
chParams = params
del chParams['stepSize']
dataflow_safe_append(df, 'Chroma2', chParams)
return df
def get_dataflow(cls, params, samplerate):
df = OnsetDetectionFunction.get_dataflow(OnsetDetectionFunction.filter_params(params), samplerate)
dataflow_safe_append(
df,
"AutoCorrelationPeaksIntegrator",
{
"NbFrames": params["BHSBeatFrameSize"],
"StepNbFrames": params["BHSBeatFrameStep"],
"ACPNbPeaks": params["ACPNbPeaks"],
"ACPNorm": "BPM",
"ACPInterPeakMinDist": "5",
},
)
dataflow_safe_append(
df,
"HistogramIntegrator",
{
"NbFrames": params["BHSHistogramFrameSize"],
"StepNbFrames": params["BHSHistogramFrameStep"],
"HInf": params["HInf"],
"HSup": params["HSup"],
"HNbBins": params["HNbBins"],
"HWeighted": "1",
},
)
dataflow_safe_append(df, "HistogramSummary", params)
return df
def get_dataflow(cls, params, samplerate):
if params["ChordsUse7"] == "1":
chtype = "maj,min,7"
else:
chtype = "maj,min"
df = Chroma.get_dataflow(
{"CQTMinFreq": "73.42", "CQTNbOctaves": "3", "CQTBinsPerOctave": "36", "stepSize": params["stepSize"]},
samplerate,
)
dataflow_safe_append(df, "Chroma2ChordDict", {"ChordTypes": chtype, "ChordNbHarmonics": "1"})
dataflow_safe_append(df, "MedianFilter", {"MFOrder": params["ChordsSmoothing"]})
dataflow_safe_append(df, "ChordDictDecoder", {"ChordTypes": chtype})
return df
def get_dataflow(cls, params, samplerate):
b = int(params["CQTBinsPerOctave"])
if b % 12 != 0:
print "WARNING: in Chroma, CQTBinsPerOctave must be multiple of 12"
b = b - b % 12
if b == 0:
b = 12
print "use CQTBinsPerOctave=%i" % b
params["CQTBinsPerOctave"] = b
df = CQT.get_dataflow(CQT.filter_params(params), samplerate)
dataflow_safe_append(df, "ChromaTune", params)
dataflow_safe_append(df, "MedianFilter", {"MFOrder": params["ChromaSmoothing"]})
dataflow_safe_append(df, "ChromaReduce", {})
return df
def get_dataflow(cls, params, samplerate):
b = int(params['CQTBinsPerOctave'])
if (b % 12 != 0):
print 'WARNING: in Chroma, CQTBinsPerOctave must be multiple of 12'
b = b - b % 12
if b == 0:
b = 12
print 'use CQTBinsPerOctave=%i' % b
params['CQTBinsPerOctave'] = b
df = CQT.get_dataflow(CQT.filter_params(params), samplerate)
dataflow_safe_append(df, 'ChromaTune', params)
dataflow_safe_append(
df, 'MedianFilter', {'MFOrder': params['ChromaSmoothing']})
dataflow_safe_append(df, 'ChromaReduce', {})
return df
def get_dataflow(cls, params, samplerate):
b = int(params['CQTBinsPerOctave'])
if (b % 12 != 0):
print('WARNING: in Chroma, CQTBinsPerOctave must be multiple of 12')
b = b - b % 12
if b == 0:
b = 12
print('use CQTBinsPerOctave=%i' % b)
params['CQTBinsPerOctave'] = b
df = CQT.get_dataflow(CQT.filter_params(params), samplerate)
dataflow_safe_append(df, 'ChromaTune', params)
dataflow_safe_append(
df, 'MedianFilter', {'MFOrder': params['ChromaSmoothing']})
dataflow_safe_append(df, 'ChromaReduce', {})
return df
def get_dataflow(cls, params, samplerate):
if (params['ChordsUse7'] == '1'):
chtype = 'maj,min,7'
else:
chtype = 'maj,min'
df = Chroma.get_dataflow({'CQTMinFreq': '73.42', 'CQTNbOctaves': '3',
'CQTBinsPerOctave': '36',
'stepSize': params['stepSize']},
samplerate)
dataflow_safe_append(df, 'Chroma2ChordDict', {'ChordTypes': chtype,
'ChordNbHarmonics': '1'})
dataflow_safe_append(df, 'MedianFilter',
{'MFOrder': params['ChordsSmoothing']})
dataflow_safe_append(df, 'ChordDictDecoder', {'ChordTypes': chtype})
return df
def get_dataflow(cls, params, samplerate):
if (params['ChordsUse7'] == '1'):
chtype = 'maj,min,7'
else:
chtype = 'maj,min'
df = Chroma.get_dataflow({'CQTMinFreq': '73.42', 'CQTNbOctaves': '3',
'CQTBinsPerOctave': '36',
'stepSize': params['stepSize']},
samplerate)
dataflow_safe_append(df, 'Chroma2ChordDict', {'ChordTypes': chtype,
'ChordNbHarmonics': '1'})
dataflow_safe_append(df, 'MedianFilter',
{'MFOrder': params['ChordsSmoothing']})
dataflow_safe_append(df, 'ChordDictDecoder', {'ChordTypes': chtype})
return df
def get_dataflow(cls, params, samplerate):
df = OnsetDetectionFunction.get_dataflow(
OnsetDetectionFunction.filter_params(params), samplerate)
dataflow_safe_append(df, 'AutoCorrelationPeaksIntegrator',
{'NbFrames': params['BHSBeatFrameSize'],
'StepNbFrames': params['BHSBeatFrameStep'],
'ACPNbPeaks': params['ACPNbPeaks'],
'ACPNorm': 'BPM',
'ACPInterPeakMinDist': '5'})
dataflow_safe_append(df, 'HistogramIntegrator',
{'NbFrames': params['BHSHistogramFrameSize'],
'StepNbFrames': params['BHSHistogramFrameStep'],
'HInf': params['HInf'],
'HSup': params['HSup'],
'HNbBins': params['HNbBins'],
'HWeighted': '1'})
dataflow_safe_append(df, 'HistogramSummary', params)
return df
def get_dataflow(cls, params, samplerate):
df = OnsetDetectionFunction.get_dataflow(
OnsetDetectionFunction.filter_params(params), samplerate)
dataflow_safe_append(df, 'AutoCorrelationPeaksIntegrator',
{'NbFrames': params['BHSBeatFrameSize'],
'StepNbFrames': params['BHSBeatFrameStep'],
'ACPNbPeaks': params['ACPNbPeaks'],
'ACPNorm': 'BPM',
'ACPInterPeakMinDist': '5'})
dataflow_safe_append(df, 'HistogramIntegrator',
{'NbFrames': params['BHSHistogramFrameSize'],
'StepNbFrames': params['BHSHistogramFrameStep'],
'HInf': params['HInf'],
'HSup': params['HSup'],
'HNbBins': params['HNbBins'],
'HWeighted': '1'})
dataflow_safe_append(df, 'HistogramSummary', params)
return df
def get_dataflow(cls, params, samplerate):
lParams = Loudness.filter_params(params)
lParams['LMode'] = 'Relative'
df = Loudness.get_dataflow(lParams, samplerate)
dataflow_safe_append(df, 'LoudnessSpread', {})
return df
def get_dataflow(cls, params, samplerate):
df = Envelope.get_dataflow(Envelope.filter_params(params), samplerate)
dataflow_safe_append(df, 'AmplitudeModulation', params)
return df
def get_dataflow(cls, params, samplerate):
df = Frames.get_dataflow(Frames.filter_params(params), samplerate)
dataflow_safe_append(df, 'Envelope', params)
return df
def get_dataflow(cls, params, samplerate):
df = DataFlow()
dataflow_safe_append(df, 'Cepstrum', params)
return df
def get_dataflow(cls, params, samplerate):
df = DataFlow()
dataflow_safe_append(df, 'AutoCorrelationPeaksIntegrator', params)
return df
def get_dataflow(cls, params, samplerate):
acparams = AutoCorrelation.filter_params(params)
acparams['ACNbCoeffs'] = str(int(params.get('LPCNbCoeffs')) + 1)
df = AutoCorrelation.get_dataflow(acparams, samplerate)
dataflow_safe_append(df, 'AC2LPC', params)
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(
MagnitudeSpectrum.filter_params(params), samplerate)
dataflow_safe_append(df, 'MelFilterBank', params)
dataflow_safe_append(df, 'Cepstrum', params)
return df
def get_dataflow(cls, params, samplerate):
df = OBSI.get_dataflow(OBSI.filter_params(params), samplerate)
dataflow_safe_append(df, 'Difference', params)
return df
def get_dataflow(cls, params, samplerate):
df = DataFlow()
dataflow_safe_append(df, 'AutoCorrelationPeaksIntegrator', params)
return df
def get_dataflow(cls, params, samplerate):
lParams = Loudness.filter_params(params)
lParams['LMode'] = 'Relative'
df = Loudness.get_dataflow(lParams, samplerate)
dataflow_safe_append(df, 'LoudnessSpread', {})
return df
def get_dataflow(cls, params, samplerate):
df = Frames.get_dataflow(Frames.filter_params(params), samplerate)
dataflow_safe_append(df, 'FFT', params)
dataflow_safe_append(df, 'ComplexDomainFlux', params)
return df
def get_dataflow(cls, params, samplerate):
df = Envelope.get_dataflow(Envelope.filter_params(params), samplerate)
dataflow_safe_append(df, 'AmplitudeModulation', params)
return df
def get_dataflow(cls, params, samplerate):
df = Envelope.get_dataflow(Envelope.filter_params(params), samplerate)
dataflow_safe_append(df, 'ShapeStatistics', params)
return df
def get_dataflow(cls, params, samplerate):
df = OBSI.get_dataflow(OBSI.filter_params(params), samplerate)
dataflow_safe_append(df, 'Difference', params)
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(
MagnitudeSpectrum.filter_params(params), samplerate)
dataflow_safe_append(df, 'NormalizeMaxAll', params)
dataflow_safe_append(
df, 'FilterSmallValues', {'FSVThreshold': '0.001'})
dataflow_safe_append(df, 'HalfHannFilter', {'HHFOrder': '0.175s'})
dataflow_safe_append(df, 'LogCompression', {})
dataflow_safe_append(
df, 'DvornikovDifferentiator', {'DDOrder': '0.08s'})
dataflow_safe_append(df, 'FilterSmallValues', {'FSVThreshold': '1'})
dataflow_safe_append(df, 'Sum', {})
dataflow_safe_append(df, 'NormalizeMaxAll', params)
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(
MagnitudeSpectrum.filter_params(params), samplerate)
dataflow_safe_append(df, 'NormalizeMaxAll', params)
dataflow_safe_append(
df, 'FilterSmallValues', {'FSVThreshold': '0.001'})
dataflow_safe_append(df, 'HalfHannFilter', {'HHFOrder': '0.175s'})
dataflow_safe_append(df, 'LogCompression', {})
dataflow_safe_append(
df, 'DvornikovDifferentiator', {'DDOrder': '0.08s'})
dataflow_safe_append(df, 'FilterSmallValues', {'FSVThreshold': '1'})
dataflow_safe_append(df, 'Sum', {})
dataflow_safe_append(df, 'NormalizeMaxAll', params)
return df
def get_dataflow(cls, params, samplerate):
df = Frames.get_dataflow(Frames.filter_params(params), samplerate)
dataflow_safe_append(df, 'AutoCorrelation', params)
return df
def get_dataflow(cls, params, samplerate):
df = Frames.get_dataflow(Frames.filter_params(params), samplerate)
dataflow_safe_append(df, 'RMS', {})
return df
def get_dataflow(cls, params, samplerate):
df = CQT.get_dataflow(params, samplerate)
dataflow_safe_append(df, 'Difference', {'DiffNbCoeffs': '0'})
return df
def get_dataflow(cls, params, samplerate):
acparams = AutoCorrelation.filter_params(params)
acparams['ACNbCoeffs'] = str(int(params.get('LPCNbCoeffs')) + 1)
df = AutoCorrelation.get_dataflow(acparams, samplerate)
dataflow_safe_append(df, 'AC2LPC', params)
return df
def get_dataflow(cls, params, samplerate):
df = Frames.get_dataflow(Frames.filter_params(params), samplerate)
dataflow_safe_append(df, 'FFT', params)
dataflow_safe_append(df, 'ComplexDomainFlux', params)
return df
def get_dataflow(cls, params, samplerate):
df = Frames.get_dataflow(Frames.filter_params(params), samplerate)
dataflow_safe_append(df, 'AutoCorrelation', params)
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(
MagnitudeSpectrum.filter_params(params), samplerate)
dataflow_safe_append(df, 'MelFilterBank', params)
dataflow_safe_append(df, 'Cepstrum', params)
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(params, samplerate)
dataflow_safe_append(df, 'Sqr', {})
dataflow_safe_append(df, 'SpectralCrestFactorPerBand', params)
return df
def get_dataflow(cls, params, samplerate):
df = Envelope.get_dataflow(Envelope.filter_params(params), samplerate)
dataflow_safe_append(df, 'ShapeStatistics', params)
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(params, samplerate)
dataflow_safe_append(df, 'Decrease', params)
return df
def get_dataflow(cls, params, samplerate):
# Power Spectrum
df = yf.MagnitudeSpectrum.get_dataflow(
yf.MagnitudeSpectrum.filter_params(params), samplerate)
dataflow_safe_append(df, 'WindowNormalize', {'NormWindow': 'Hanning'})
dataflow_safe_append(df, 'Sqr', {})
#dataflow_safe_append(df,'DCOffsetFilter',{}) #only good for special files
dataflow_safe_append(df, 'DBConversion', {})
# Emphasize Local Peaks
dataflow_safe_append(df, 'SubRunningAverage', params)
# Binarization
dataflow_safe_append(df, 'Binarization', params)
# Calculate Frequency Activation
dataflow_safe_append(df, 'FrameSum', params)
# Detect Strong Peaks
dataflow_safe_append(df, 'PeakDetection', params)
# Quantify CFA
dataflow_safe_append(df, 'Sum', params)
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(params, samplerate)
dataflow_safe_append(df, 'ShapeStatistics', params)
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(
MagnitudeSpectrum.filter_params(params), samplerate)
dataflow_safe_append(df, 'Sqr', {})
dataflow_safe_append(df, 'OBSI', params)
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(params, samplerate)
dataflow_safe_append(df, 'Sqr', {})
dataflow_safe_append(df, 'Rolloff', params)
return df
def get_dataflow(cls, params, samplerate):
df = Frames.get_dataflow(Frames.filter_params(params), samplerate)
dataflow_safe_append(df, 'ShapeStatistics', {})
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(params, samplerate)
dataflow_safe_append(df, 'Sqr', {})
dataflow_safe_append(df, 'SpectralCrestFactorPerBand', params)
return df
def get_dataflow(cls, params, samplerate):
df = MagnitudeSpectrum.get_dataflow(
MagnitudeSpectrum.filter_params(params), samplerate)
dataflow_safe_append(df, 'Sqr', {})
dataflow_safe_append(df, 'OBSI', params)
return df
def get_dataflow(cls, params, samplerate):
df = DataFlow()
dataflow_safe_append(df, 'Cepstrum', params)
return df
