def block_mix_wavs(wavpath_a,
wavpath_b,
out_wavpath,
a_gain=1.,
b_gain=1.,
block_size=4096,
mute_left=False):
'''
Mix two wav files, applying gains to each
'''
wav_a = PySndfile(wavpath_a, 'r')
wav_b = PySndfile(wavpath_b, 'r')
out_wav = PySndfile(out_wavpath, 'w', construct_format('wav', 'pcm16'),
wav_a.channels(), wav_a.samplerate())
i = 0
while i < wav_a.frames():
if i + block_size > wav_a.frames():
block_size = wav_a.frames() - i
x1 = wav_a.read_frames(block_size)
x2 = wav_b.read_frames(block_size)
x1[:, :2] *= a_gain
x2 *= b_gain
if x1.shape[1] == 3:
y = np.zeros(x1.shape)
y[:, 0] = x1[:, 0] + x2
y[:, 1] = x1[:, 1] + x2
y[:, 2] = x1[:, 2]
if mute_left:
y[:, 0] = 0.0
else:
y = x1 + x2
out_wav.write_frames(y)
i += block_size
def block_process_wav(wavpath, out_wavpath, func, block_size=4096, **args):
'''
Mix two wav files, applying gains to each
'''
wav = PySndfile(wavpath, 'r')
out_wav = PySndfile(out_wavpath, 'w', construct_format('wav', 'pcm16'),
wav.channels(), wav.samplerate())
i = 0
while i < wav.frames():
if i + block_size > wav.frames():
block_size = wav.frames() - i
x = wav.read_frames(block_size)
y = func(x, **args)
out_wav.write_frames(y)
i += block_size
del out_wav
def __init__(self, fn, sr=None, chns=None, blksz=2**16, dtype=np.float32):
fnd = False
if not fnd and (PySndfile is not None):
try:
sf = PySndfile(fn, mode='r')
except IOError:
pass
else:
if (sr is None or sr == sf.samplerate()) and (
chns is None or chns == sf.channels()):
# no resampling required
self.channels = sf.channels()
self.samplerate = sf.samplerate()
self.frames = sf.frames()
self.rdr = sndreader(sf, blksz, dtype=dtype)
fnd = True
if not fnd:
ffmpeg = findfile('ffmpeg') or findfile('avconv')
if ffmpeg is not None:
pipe = sp.Popen([ffmpeg, '-i', fn, '-'],
stdin=sp.PIPE,
stdout=sp.PIPE,
stderr=sp.PIPE)
fmtout = pipe.stderr.read()
if (sys.version_info > (3, 0)):
fmtout = fmtout.decode()
m = re.match(
r"^(ffmpeg|avconv) version.*Duration: (\d\d:\d\d:\d\d.\d\d),.*Audio: (.+), (\d+) Hz, (.+), (.+), (\d+) kb/s",
" ".join(fmtout.split('\n')))
if m is not None:
self.samplerate = int(m.group(4)) if not sr else int(sr)
chdef = m.group(5)
if chdef.endswith(" channels") and len(chdef.split()) == 2:
self.channels = int(chdef.split()[0])
else:
try:
self.channels = {
'mono': 1,
'1 channels (FL+FR)': 1,
'stereo': 2,
'hexadecagonal': 16
}[chdef] if not chns else chns
except:
print(f"Channel definition '{chdef}' unknown")
raise
dur = reduce(lambda x, y: x * 60 + y,
list(map(float,
m.group(2).split(':'))))
self.frames = int(
dur * self.samplerate
) # that's actually an estimation, because of potential resampling with round-off errors
pipe = sp.Popen(
[
ffmpeg, '-i', fn, '-f', 'f32le', '-acodec',
'pcm_f32le', '-ar',
str(self.samplerate), '-ac',
str(self.channels), '-'
],
# bufsize=self.samplerate*self.channels*4*50,
stdin=sp.PIPE,
stdout=sp.PIPE,
stderr=sp.PIPE)
def rdr():
bufsz = (blksz // self.channels) * self.channels * 4
while True:
data = pipe.stdout.read(bufsz)
if len(data) == 0:
break
data = np.fromstring(data, dtype=dtype)
yield data.reshape((-1, self.channels)).T
self.rdr = rdr()
fnd = True
if not fnd:
raise IOError("Format not usable")