1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181
|
"""
Module to read / write wav files using numpy arrays
Functions
---------
`read`: Return the sample rate (in samples/sec) and data from a WAV file.
`write`: Write a numpy array as a WAV file.
"""
import numpy
from numpy.compat import asbytes
import struct
import warnings
class WavFileWarning(UserWarning):
pass
_big_endian = False
# assumes file pointer is immediately
# after the 'fmt ' id
def _read_fmt_chunk(fid):
if _big_endian:
fmt = '>'
else:
fmt = '<'
res = struct.unpack(fmt+'ihHIIHH',fid.read(20))
size, comp, noc, rate, sbytes, ba, bits = res
if (comp != 1 or size > 16):
warnings.warn("Unfamiliar format bytes", WavFileWarning)
if (size>16):
fid.read(size-16)
return size, comp, noc, rate, sbytes, ba, bits
# assumes file pointer is immediately
# after the 'data' id
def _read_data_chunk(fid, noc, bits):
if _big_endian:
fmt = '>i'
else:
fmt = '<i'
size = struct.unpack(fmt,fid.read(4))[0]
if bits == 8:
data = numpy.fromfile(fid, dtype=numpy.ubyte, count=size)
if noc > 1:
data = data.reshape(-1,noc)
else:
bytes = bits//8
if _big_endian:
dtype = '>i%d' % bytes
else:
dtype = '<i%d' % bytes
data = numpy.fromfile(fid, dtype=dtype, count=size//bytes)
if noc > 1:
data = data.reshape(-1,noc)
return data
def _read_riff_chunk(fid):
global _big_endian
str1 = fid.read(4)
if str1 == asbytes('RIFX'):
_big_endian = True
elif str1 != asbytes('RIFF'):
raise ValueError("Not a WAV file.")
if _big_endian:
fmt = '>I'
else:
fmt = '<I'
fsize = struct.unpack(fmt, fid.read(4))[0] + 8
str2 = fid.read(4)
if (str2 != asbytes('WAVE')):
raise ValueError("Not a WAV file.")
if str1 == asbytes('RIFX'):
_big_endian = True
return fsize
# open a wave-file
def read(file):
"""
Return the sample rate (in samples/sec) and data from a WAV file
Parameters
----------
file : file
Input wav file.
Returns
-------
rate : int
Sample rate of wav file
data : numpy array
Data read from wav file
Notes
-----
* The file can be an open file or a filename.
* The returned sample rate is a Python integer
* The data is returned as a numpy array with a
data-type determined from the file.
"""
if hasattr(file,'read'):
fid = file
else:
fid = open(file, 'rb')
fsize = _read_riff_chunk(fid)
noc = 1
bits = 8
while (fid.tell() < fsize):
# read the next chunk
chunk_id = fid.read(4)
if chunk_id == asbytes('fmt '):
size, comp, noc, rate, sbytes, ba, bits = _read_fmt_chunk(fid)
elif chunk_id == asbytes('data'):
data = _read_data_chunk(fid, noc, bits)
else:
warnings.warn("chunk not understood", WavFileWarning)
data = fid.read(4)
if _big_endian:
fmt = '>i'
else:
fmt = '<i'
size = struct.unpack(fmt, data)[0]
fid.seek(size, 1)
fid.close()
return rate, data
# Write a wave-file
# sample rate, data
def write(filename, rate, data):
"""
Write a numpy array as a WAV file
Parameters
----------
filename : file
The name of the file to write (will be over-written).
rate : int
The sample rate (in samples/sec).
data : ndarray
A 1-D or 2-D numpy array of integer data-type.
Notes
-----
* Writes a simple uncompressed WAV file.
* The bits-per-sample will be determined by the data-type.
* To write multiple-channels, use a 2-D array of shape
(Nsamples, Nchannels).
"""
fid = open(filename, 'wb')
fid.write(asbytes('RIFF'))
fid.write(asbytes('\x00\x00\x00\x00'))
fid.write(asbytes('WAVE'))
# fmt chunk
fid.write(asbytes('fmt '))
if data.ndim == 1:
noc = 1
else:
noc = data.shape[1]
bits = data.dtype.itemsize * 8
sbytes = rate*(bits // 8)*noc
ba = noc * (bits // 8)
fid.write(struct.pack('<ihHIIHH', 16, 1, noc, rate, sbytes, ba, bits))
# data chunk
fid.write(asbytes('data'))
fid.write(struct.pack('<i', data.nbytes))
import sys
if data.dtype.byteorder == '>' or (data.dtype.byteorder == '=' and sys.byteorder == 'big'):
data = data.byteswap()
data.tofile(fid)
# Determine file size and place it in correct
# position at start of the file.
size = fid.tell()
fid.seek(4)
fid.write(struct.pack('<i', size-8))
fid.close()
|