view Example VamPy plugins/PySpectralFeatures.py @ 24:7d28bed0864e

* Rearrange Python plugin construction.

  Formerly, the PyPluginAdapter has retained a single plugin instance pointer
  for each plugin found, and its createPlugin method has simply returned a new
  PyPlugin object wrapping the same instance pointer.  This has a couple of
  negative consequences:
   - Because construction of the actual Python instance occurred before the
     wrapper was constructed, it was not possible to pass arguments (i.e.
     the sample rate) from the wrapper constructor to the Python plugin
     instance constructor -- they had to be passed later, to initialise,
     disadvantaging those plugins that would like to use the sample rate
     for parameter & step/block size calculations etc
   - Because there was only a single Python plugin instance, it was not
     possible to run more than one instance at once with any isolation

  This rework instead stores the Python class pointer (rather than instance
  pointer) in the PyPluginAdapter, and each PyPlugin wrapper instance creates
  its own Python plugin instance.  What could possibly go wrong?

'''PySpectralFeatures.py - Example plugin demonstrates''' 
'''how to use the NumPy array interface and write Matlab style code.'''

from numpy import *

class PySpectralFeatures: 
	
	def __init__(self,inputSampleRate): 
		self.m_inputSampleRate = inputSampleRate
		self.m_stepSize = 0
		self.m_blockSize = 0
		self.m_channels = 0
		self.threshold = 0.00
		self.r = 2.0
		
	def initialise(self,channels,stepSize,blockSize):
		self.m_channels = channels
		self.m_stepSize = stepSize		
		self.m_blockSize = blockSize
		#self.prevMag = ones((blockSize/2)-1) / ((blockSize/2)-1)
		self.prevMag = zeros((blockSize/2)-1)
		self.prevMag[0] = 1

		return True
	
	def getMaker(self):
		return 'VamPy Example Plugins'
	
	def getName(self):
		return 'VamPy Spectral Features'
		
	def getIdentifier(self):
		return 'vampy-sf2'

	def getDescription(self):
		return 'A collection of low-level spectral descriptors.'
	
	def getMaxChannelCount(self):
		return 1
		
	def getInputDomain(self):
		return 'FrequencyDomain'
			
	def getOutputDescriptors(self):

		#descriptors are python dictionaries
		#Generic values are the same for all
		Generic={
		'hasFixedBinCount':True,
		'binCount':1,
		'hasKnownExtents':False,
		'isQuantized':False,
		'sampleType':'OneSamplePerStep'
		}

		#Spectral centroid etc...
		SC=Generic.copy()
		SC.update({
		'identifier':'vampy-sc',
		'name':'Spectral Centroid',
		'description':'Spectral Centroid (Brightness)',
		'unit':'Hz'		
		})
				
		SCF=Generic.copy()
		SCF.update({
		'identifier':'vampy-scf',
		'name':'Spectral Crest Factor',
		'description':'Spectral Crest (Tonality)',
		'unit':'v'
		})

		BW=Generic.copy()
		BW.update({
		'identifier':'vampy-bw',
		'name':'Band Width',
		'description':'Spectral Band Width',
		'unit':'Hz',
		})		
		
		SE=Generic.copy()
		SE.update({
		'identifier':'vampy-se',
		'name':'Shannon Entropy',
		'description':'Shannon Entropy',
		'unit':'',
		})		

		RE=Generic.copy()
		RE.update({
		'identifier':'vampy-re',
		'name':'Renyi Entropy',
		'description':'Renyi Entropy',
		'unit':'',
		})

		KL=Generic.copy()
		KL.update({
		'identifier':'vampy-kl',
		'name':'Kullback Leibler divergence',
		'description':'KL divergence between successive spectra',
		'unit':'',
		})
			
		#return a list of dictionaries
		return [SC,SCF,BW,SE,RE,KL]

	def getParameterDescriptors(self):
		threshold={
		'identifier':'threshold',
		'name':'Noise threshold: ',
		'description':'',
		'unit':'v',
		'minValue':0.0,
		'maxValue':0.5,
		'defaultValue':0.05,
		'isQuantized':False
		}

		renyicoeff={
		'identifier':'r',
		'name':'Renyi entropy coeff: ',
		'description':'',
		'unit':'',
		'minValue':0.0,
		'maxValue':10.0,
		'defaultValue':2,
		'isQuantized':False		
		}

		return [threshold,renyicoeff]

	def setParameter(self,paramid,newval):
		if paramid == 'threshold' :
			self.threshold = newval
		if paramid == 'r' :
			self.r == newval
		return
		
	def getParameter(self,paramid):
		if paramid == 'threshold' :
			return self.threshold
		if paramid == 'r':
			return float(self.r)
		else:
			return 0.0
			
	def processN(self,membuffer,samplecount):
		fftsize = self.m_blockSize
		sampleRate = self.m_inputSampleRate

		#for time domain plugins use the following line:
		#audioSamples = frombuffer(membuffer[0],float32)
		#-1: do till the end, skip DC 2*32bit / 8bit = 8byte
		complexSpectrum =  frombuffer(membuffer[0],complex64,-1,8)
		magnitudeSpectrum = abs(complexSpectrum) / (fftsize/2)
		tpower = sum(magnitudeSpectrum)
		#phaseSpectrum = angle(complexSpectrum)

		freq = array(range(1,len(complexSpectrum)+1)) \
		* sampleRate / fftsize

		centroid = 0.0
		crest = 0.0
		bw = 0.0
		shannon = 0.0
		renyi = 0.0
		r = self.r
		KLdiv = 0.0
		flatness = 0.0
		exp=1.0 / (fftsize/2)
		#print exp

		#declare outputs
		output0=[]
		output1=[]
		output2=[]
		output3=[]
		output4=[]
		output5=[]
		
		if tpower > self.threshold : 

			centroid = sum(freq * magnitudeSpectrum) / tpower 
			crest = max(magnitudeSpectrum)  / tpower
			bw = sum( abs(freq - centroid) * magnitudeSpectrum ) / tpower
			#flatness = prod(abs(complexSpectrum))  
			#print flatness
			normMag = magnitudeSpectrum / tpower #make it sum to 1			
			shannon = - sum( normMag * log2(normMag) )
			renyi = (1/1-r) * log10( sum( power(normMag,r)))
			KLdiv = sum( normMag * log2(normMag / self.prevMag) )
			self.prevMag = normMag
 				
		output0.append({
		'hasTimestamp':False,		
		'values':[float(centroid)],		
		#'label':str(centroid)				
		})

		output1.append({
		'hasTimestamp':False,		
		'values':[float(crest)],		
		#'label':str(crest)				
		})
	
		output2.append({
		'hasTimestamp':False,		
		'values':[float(bw)],		
		#'label':str(bw)				
		})

		output3.append({
		'hasTimestamp':False,		
		'values':[float(shannon)],		
		#'label':str(shannon)				
		})

		output4.append({
		'hasTimestamp':False,		
		'values':[float(renyi)],		
		#'label':str(renyi)				
		})

		output5.append({
		'hasTimestamp':False,		
		'values':[float(KLdiv)],		#strictly must be a list
		#'label':str(renyi)				
		})

		#return a LIST of list of dictionaries
		return [output0,output1,output2,output3,output4,output5]