def ignorm(c, gamma):
"""
gnorm performs inverse gain normalization on generalized cepstrum
Parameters
----------
c : array, shape (`order of cepstrum` + 1)
generalized cepstrum
gamma : float
TODO
Return
------
generalized cepstrum : array, shape(=c)
"""
c2 = np.zeros(c.shape)
csptk.ignorm(c, c2, gamma)
return c2
def mgcep(x, order=20, alpha=0.41, gamma=0.0, n=None,
iter1=2, iter2=30,
dd=0.001, etype=0, e=0.0, f=0.0001, itype=0, otype=0):
"""
Mel Generalized Cepstrum analysis
Parameters
----------
x : array, shape (`frame_len`)
A input frame
order : int
order of mel generalized cepstrum that will be extracted
alpha : float
all pass constant
n : int (default=len(x)-1)
order of recursions
iter1 : int
minimum number of iteration
iter2 : int
maximum number of iteration
dd : float
threshold
etype : int
type of paramter `e`
(0) not used
(1) initial value of log-periodogram
(2) floor of periodogram in db
e : float
initial value for log-periodogram or floor of periodogram in db
f : float
mimimum value of the determinant of normal matrix
itype : float
input data type:
(0) windowed signal
(1) log amplitude in db
(2) log amplitude
(3) amplitude
(4) periodogram
otype : int
output data type
(0) mel generalized cepstrum: (c~0...c~m)
(1) MGLSA filter coefficients: b0...bm
(2) K~,c~'1...c~'m
(3) K,b'1...b'm
(4) K~,g*c~'1...g*c~'m
(5) K,g*b'1...g*b'm
Return
------
mel generalized cepstrum : array, shape (`order + 1`)
"""
mgc = np.zeros(order+1)
if n == None:
n = len(x)-1
end_condition = csptk.mgcep(x, mgc, alpha, gamma, n,
iter1, iter2,
dd, etype, e, f, itype)
if otype == 0 or otype == 1 or otype == 2 or otype == 4:
csptk.ignorm(mgc, mgc, gamma)
if otype == 0 or otype == 2 or otype == 4:
csptk.b2mc(mgc, mgc, alpha)
if otype == 2 or otype == 4:
csptk.gnorm(mgc, mgc, gamma)
if otype == 4 or otype == 5:
mgc[1:] *= gamma
return mgc
