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(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 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)
parser.error("Input file '%s' not found"%args.input)
fs = args.sr
# build transform
scales = {'log':LogScale, 'lin':LinScale, 'mel':MelScale, 'oct':OctScale}
try:
scale = scales[args.scale]
except KeyError:
parser.error('Scale unknown (--scale option)')
scl = scale(args.fmin, args.fmax, args.bins, beyond=int(args.reducedform == 2))
slicq = NSGT_sliced(scl, args.sllen, args.trlen, fs,
real=args.real, recwnd=args.recwnd,
matrixform=args.matrixform, reducedform=args.reducedform,
multithreading=args.multithreading,
multichannel=True
)
# Read audio data
sf = SndReader(args.input, sr=fs, chns=2)
signal = sf()
# duration of signal in s
dur = sf.frames/float(fs)
# total number of coefficients to represent input signal
ncoefs = int(sf.frames*slicq.coef_factor)
# read slices from audio file and mix down signal, if necessary at all
if not args.downmix_after:
fs = args.sr
# build transform
scales = {'log': LogScale, 'lin': LinScale, 'mel': MelScale, 'oct': OctScale}
try:
scale = scales[args.scale]
except KeyError:
parser.error('Scale unknown (--scale option)')
scl = scale(args.fmin, args.fmax, args.bins, beyond=int(args.reducedform == 2))
slicq = NSGT_sliced(scl,
args.sllen,
args.trlen,
fs,
real=args.real,
recwnd=args.recwnd,
matrixform=args.matrixform,
reducedform=args.reducedform,
multithreading=args.multithreading,
multichannel=True)
# Read audio data
sf = SndReader(args.input, sr=fs, chns=2)
signal = sf()
# duration of signal in s
dur = sf.frames / float(fs)
# total number of coefficients to represent input signal
ncoefs = int(sf.frames * slicq.coef_factor)
signal = torch.cat(signal, dim=-1)
# duration of signal in s
dur = sf.frames / float(fs)
if args.sllen is None:
sllen, trlen = scl.suggested_sllen_trlen(fs)
else:
sllen = args.sllen
trlen = args.trlen
if not args.nonsliced:
slicq = NSGT_sliced(scl,
sllen,
trlen,
fs,
real=True,
matrixform=True,
multichannel=True,
device="cpu")
else:
slicq = NSGT(scl,
fs,
signal.shape[-1],
real=True,
matrixform=True,
multichannel=True,
device="cpu")
# total number of coefficients to represent input signal
#ncoefs = int(sf.frames*slicq.coef_factor)
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
if sf.channels > 1:
s = np.mean(s, axis=1)
if args.length:
s = s[:args.length]
scales = {'log':LogScale,'lin':LinScale,'mel':MelScale,'oct':OctScale}
try:
scale = scales[args.scale]
except KeyError:
parser.error('scale unknown')
scl = scale(args.fmin, args.fmax, args.bins)
slicq = NSGT_sliced(scl, args.sllen, args.trlen, fs,
real=args.real, recwnd=args.recwnd,
matrixform=args.matrixform, reducedform=args.reducedform,
multithreading=args.multithreading
)
t1 = cputime()
signal = (s,)
# generator for forward transformation
c = slicq.forward(signal)
# realize transform from generator
c = list(c)
# generator for backward transformation
outseq = slicq.backward(c)
fs = sf.samplerate
s = sf.read_frames(sf.nframes)
if sf.channels > 1:
s = N.mean(s,axis=1)
if options.length:
s = s[:options.length]
scales = {'log':LogScale,'lin':LinScale,'mel':MelScale,'oct':OctScale}
try:
scale = scales[options.scale]
except KeyError:
parser.error('scale unknown')
scl = scale(options.fmin,options.fmax,options.bins)
slicq = NSGT_sliced(scl,options.sl_len,options.tr_area,fs,real=options.real,recwnd=options.recwnd,matrixform=options.matrixform,reducedform=options.reducedform)
t1 = time()
signal = (s,)
# generator for forward transformation
c = slicq.forward(signal)
# realize transform from generator
c = list(c)
# cl = map(len,c[0])
# print "c",len(cl),cl
# generator for backward transformation
scales = {
'log': LogScale,
'lin': LinScale,
'mel': MelScale,
'oct': OctScale
}
try:
scale = scales[options.scale]
except KeyError:
parser.error('scale unknown')
scl = scale(options.fmin, options.fmax, options.bins)
slicq = NSGT_sliced(scl,
options.sl_len,
options.tr_area,
fs,
real=options.real,
recwnd=options.recwnd,
matrixform=options.matrixform,
reducedform=options.reducedform)
t1 = time()
signal = (s, )
# generator for forward transformation
c = slicq.forward(signal)
# realize transform from generator
c = list(c)
# cl = map(len,c[0])
except KeyError:
parser.error('Scale unknown (--scale option)')
scl = scale(args.fmin, args.fmax, args.bins)
if args.sllen is None:
sllen, trlen = scl.suggested_sllen_trlen(fs)
else:
sllen = args.sllen
trlen = args.trlen
print(f'sllen: {sllen}, trlen: {trlen}')
slicq = NSGT_sliced(scl,
sllen,
trlen,
fs,
real=True,
matrixform=args.matrixform,
multichannel=True,
device="cpu")
# Read audio data
sf = SndReader(args.input, sr=fs, chns=2)
signal = sf()
signal = [torch.tensor(sig) for sig in signal]
#pad = signal[0].shape[-1]-signal[-1].shape[-1]
#signal[-1] = torch.nn.functional.pad(signal[-1], (0, pad), mode='constant', value=0)
signal = torch.cat(signal, dim=-1)
# duration of signal in s
slicq = NSGT_sliced_old(scl,
sllen,
trlen,
fs,
real=True,
matrixform=args.matrixform,
multithreading=args.multithreading,
multichannel=True)
# read slices from audio file and mix down signal, if necessary at all
signal = ((np.mean(s, axis=0), ) for s in signal_orig)
else:
slicq = NSGT_sliced(scl,
sllen,
trlen,
fs,
real=True,
matrixform=args.matrixform,
multichannel=True,
device=args.torch_device)
signal = [
torch.tensor(sig, device=args.torch_device) for sig in signal_orig
]
pad = signal[0].shape[-1] - signal[-1].shape[-1]
signal[-1] = torch.nn.functional.pad(signal[-1], (0, pad),
mode='constant',
value=0)
signal = torch.cat(signal, dim=-1)
tot = 0.
