Edit: here is an updated version of scipy.io.wavfile
that adds many things (24 bit .wav files support for read/write, cue markers, cue markers labels, and some other metadata like pitch (if defined), etc.):
wavfile.py (enhanced)
Feel free to share it!
I finally found a solution (it uses some function of scipy.io.wavfile) :
def readmarkers(file, mmap=False):
if hasattr(file,'read'):
fid = file
else:
fid = open(file, 'rb')
fsize = _read_riff_chunk(fid)
cue = []
while (fid.tell() < fsize):
chunk_id = fid.read(4)
if chunk_id == b'cue ':
size, numcue = struct.unpack('<ii',fid.read(8))
for c in range(numcue):
id, position, datachunkid, chunkstart, blockstart, sampleoffset = struct.unpack('<iiiiii',fid.read(24))
cue.append(position)
else:
_skip_unknown_chunk(fid)
fid.close()
return cue
Feel free to add it into Scipy's wavfile.py
if someone is interested.