def test_oct(self):
siglen = int(10**np.random.uniform(4, 6))
sig = np.random.random(siglen)
fmin = np.random.random() * 200 + 20
fmax = np.random.random() * (22048 - fmin) + fmin
obins = np.random.randint(24) + 1
scale = OctScale(fmin, fmax, obins)
nsgt = NSGT(scale, fs=44100, Ls=len(sig))
c = nsgt.forward(sig)
s_r = nsgt.backward(c)
self.assertTrue(np.allclose(sig, s_r, atol=1e-07))
def getiNSGT(C, L, Fs, resol=24):
"""
Perform an inverse Nonstationary Gabor Transform
:param C: An NBinsxNFrames CQT array
:param L: Number of samples in audio file
:param Fs: Sample rate
:param resol: Number of CQT bins per octave
"""
from nsgt import NSGT, OctScale
scl = OctScale(50, Fs, resol)
nsgt = NSGT(scl, Fs, L, matrixform=True)
return nsgt.backward(C)
def getNSGT(X, Fs, resol=24):
"""
Perform a Nonstationary Gabor Transform implementation of CQT
:param X: A 1D array of audio samples
:param Fs: Sample rate
:param resol: Number of CQT bins per octave
"""
from nsgt import NSGT, OctScale
scl = OctScale(50, Fs, resol)
nsgt = NSGT(scl, Fs, len(X), matrixform=True)
C = nsgt.forward(X)
return np.array(C)
def runit(self, siglen, fmin, fmax, obins, sllen, trlen, real):
sig = rndsig[:siglen]
scale = OctScale(fmin, fmax, obins)
nsgt = NSGT_sliced(scale, fs=44100, sl_len=sllen, tr_area=trlen, real=real)
c = nsgt.forward((sig,))
rc = nsgt.backward(c)
s_r = np.concatenate(list(map(list,rc)))[:len(sig)]
close = np.allclose(sig, s_r, atol=1.e-3)
if not close:
print("Failing params:", siglen, fmin, fmax, obins, sllen, trlen, real)
dev = np.abs(s_r-sig)
print("Error", np.where(dev>1.e-3), np.max(dev))
self.assertTrue(close)
def getiNSGTGriffinLim(C, L, Fs, resol=24, randPhase=False, NIters=20):
from nsgt import NSGT, OctScale
scl = OctScale(50, Fs, resol)
nsgt = NSGT(scl, Fs, L, matrixform=True)
eps = 2.2204e-16
if randPhase:
C = np.exp(
np.complex(0, 1) * np.random.rand(C.shape[0], C.shape[1])) * C
A = np.array(C, dtype=np.complex)
for i in range(NIters):
print("iNSGT Griffin Lim Iteration %i of %i" % (i + 1, NIters))
Ai = np.array(nsgt.forward(nsgt.backward(C)))
A = np.zeros_like(C)
A[:, 0:Ai.shape[1]] = Ai
Norm = np.sqrt(A * np.conj(A))
Norm[Norm < eps] = 1
A = np.abs(C) * (A / Norm)
X = nsgt.backward(A)
return np.real(X)
def main():
parser = ArgumentParser()
parser.add_argument(
"--mask",
type=str,
default="soft",
choices=("hard", "soft"),
help="mask strategy",
)
parser.add_argument("--outdir", type=str, default="./", help="output directory")
parser.add_argument(
"--stream-size",
type=int,
default=1024,
help="stream size for simulated realtime from wav (default=%(default)s)",
)
parser.add_argument("input", type=str, help="input file")
args = parser.parse_args()
prefix = args.input.split("/")[-1].split("_")[0]
harm_out = os.path.join(args.outdir, prefix + "_harmonic.wav")
perc_out = os.path.join(args.outdir, prefix + "_percussive.wav")
print("writing files to {0}, {1}".format(harm_out, perc_out))
lharm = 17
lperc = 7
# calculate transform parameters
nsgt_scale = OctScale(80, 20000, 12)
trlen = args.stream_size # transition length
sllen = 4 * args.stream_size # slice length
x, fs = librosa.load(args.input, sr=None)
xh = numpy.zeros_like(x)
xp = numpy.zeros_like(x)
hop = trlen
chunk_size = hop
n_chunks = int(numpy.floor(x.shape[0] // hop))
eps = numpy.finfo(numpy.float32).eps
slicq = NSGT_sliced(
nsgt_scale,
sllen,
trlen,
fs,
real=True,
matrixform=True,
)
total_time = 0.0
for chunk in range(n_chunks - 1):
t1 = cputime()
start = chunk * hop
end = start + sllen
s = x[start:end]
signal = (s,)
c = slicq.forward(signal)
c = list(c)
C = numpy.asarray(c)
Cmag = numpy.abs(C)
H = scipy.ndimage.median_filter(Cmag, size=(1, lharm, 1))
P = scipy.ndimage.median_filter(Cmag, size=(1, 1, lperc))
if args.mask == "soft":
# soft mask first
tot = numpy.power(H, 2.0) + numpy.power(P, 2.0) + eps
Mp = numpy.divide(numpy.power(H, 2.0), tot)
Mh = numpy.divide(numpy.power(P, 2.0), tot)
else:
Mh = numpy.divide(H, P + eps) > 2.0
Mp = numpy.divide(P, H + eps) >= 2.0
Cp = numpy.multiply(Mp, C)
Ch = numpy.multiply(Mh, C)
# generator for backward transformation
outseq_h = slicq.backward(Ch)
outseq_p = slicq.backward(Cp)
# make single output array from iterator
sh_r = next(reblock(outseq_h, len(s), fulllast=False))
sh_r = sh_r.real
sp_r = next(reblock(outseq_p, len(s), fulllast=False))
sp_r = sp_r.real
xh[start:end] = sh_r
xp[start:end] = sp_r
t2 = cputime()
total_time += t2 - t1
print("Calculation time per iter: %fs" % (total_time / n_chunks))
scipy.io.wavfile.write(harm_out, fs, xh)
scipy.io.wavfile.write(perc_out, fs, xp)
return 0