Reply by david bateman May 28, 20072007-05-28
On May 27, 5:05 am, Randy Yates <y...@ieee.org> wrote:
> I wrote: > > I'm using octave 2.9.9 on > > > [yates@localhost inp]$ uname -a > > Linux localhost.localdomain 2.6.20-1.2948.fc6 #1 SMP Fri Apr 27 19:18:54 EDT 2007 x86_64 x86_64 x86_64 GNU/Linux > > > and I'm experiencing two problems with wavwrite(): > > > 1. The parameter order changed between version 2.9.8 and 2.9.9. This is > > not really a problem, but may be noteworthy. > > > 2. wavwrite() doesn't seem to be able to create a valid .wav file: > > > [yates] octave > > >x=2*rand(1,128)-1; > > >wavwrite('test.wav',x,64000,16); > > >quit > > > [yates] play test.wav > > sox: Failed reading test.wav: WAVE header not found > > > What's up with this? > > The problem is caused by bugs in wavwrite.m. Namely, it assumes > lengths for the architecture-dependent data types "short" and "ulong". > I'm including an updated version that is corrected and seems to work. > > PS: I believe you have the same problem in ausave() since it gave > the exact same error. > -- > % Randy Yates % "And all that I can do > %% Fuquay-Varina, NC % is say I'm sorry, > %%% 919-577-9882 % that's the way it goes..." > %%%% <y...@ieee.org> % Getting To The Point', *Balance of Power*, ELOhttp://home.earthlink.net/~yatescr > > ## Copyright (C) 2005 Michael Zeising > ## > ## This file is part of Octave. > ## > ## Octave is free software; you can redistribute it and/or modify it > ## under the terms of the GNU General Public License as published by > ## the Free Software Foundation; either version 2, or (at your option) > ## any later version. > ## > ## Octave is distributed in the hope that it will be useful, but > ## WITHOUT ANY WARRANTY; without even the implied warranty of > ## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > ## General Public License for more details. > ## > ## You should have received a copy of the GNU General Public License > ## along with Octave; see the file COPYING. If not, write to the Free > ## Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA > ## 02110-1301, USA. > > ## -*- texinfo -*- > ## @deftypefn {Function File} {} wavwrite(@var{filename}, @var{y}) > ## Write @var{y} to the canonical RIFF/WAVE sound file @var{filename}. A sample > ## rate of 8000 Hz and 16-bit samples are assumed. Each column of the data > ## represents a separate channel. > ## > ## @deftypefnx {Function File} {} wavwrite(@var{filename}, @var{y}, @var{fs}) > ## Set the sample rate to @var{fs} Hz. > ## > ## @deftypefnx {Function File} {} wavwrite(@var{filename}, @var{y}, @var{fs}, @var{bits}) > ## Set the sample rate to @var{fs} Hz and resolution to @var{bits} bits. > ## @seealso{wavread} > ## @end deftypefn > > ## Author: Michael Zeising <michael.zeis...@stud.uni-erlangen.de> > ## Created: 06 December 2005 > > function wavwrite (filename, y, samples_per_sec, bits_per_sample) > > BYTEORDER = "ieee-le"; > > if (nargin < 2 || nargin > 4) > usage ("wavwrite (filename, y, samples_per_sec, bits_per_sample)"); > endif > > ## parse arguments > if (nargin < 3) > warning ("wavwrite: sample rate set to 8000 Hz"); > samples_per_sec = 8000; > endif > > if (nargin < 4) > warning ("wavwrite: sample resolution set to 16-bit"); > bits_per_sample = 16; > endif > > ## determine sample format > switch (bits_per_sample) > case 8 > format = "uint8"; > case 16 > format = "int16"; > case 32 > format = "int32"; > otherwise > error ("wavread: sample resolution not supported"); > endswitch > > ## calculate filesize > channels = size(y)(2); > n = size(y)(1); > > ## size of data chunk > ck_size = n*channels*(bits_per_sample/8); > > ## open file for writing binary > > if (! ischar (filename)) > error ("wavwrite: expecting filename to be a character string"); > endif > > [fid, msg] = fopen (filename, "wb"); > if (fid < 0) > error ("wavwrite: %s", msg) > endif > > ## write RIFF/WAVE header > c = 0; > c += fwrite (fid, "RIFF", "uchar"); > > ## file size - 8 > c += fwrite (fid, ck_size + 36, "uint32", 0, BYTEORDER); > c += fwrite (fid, "WAVEfmt ", "uchar"); > > ## size of fmt chunk > c += fwrite (fid, 16, "uint32", 0, BYTEORDER); > > ## sample format code (PCM) > c += fwrite (fid, 0x0001, "uint16", 0, BYTEORDER); > > ## channels > c += fwrite (fid, channels, "uint16", 0, BYTEORDER); > > ## sample rate > c += fwrite (fid, samples_per_sec, "uint32", 0, BYTEORDER); > > ## bytes per second > bps = samples_per_sec*channels*bits_per_sample/8; > c += fwrite (fid, bps, "uint32", 0, BYTEORDER); > > ## block align > c += fwrite (fid, channels*bits_per_sample/8, "uint16", 0, BYTEORDER); > > c += fwrite (fid, bits_per_sample, "uint16", 0, BYTEORDER); > c += fwrite (fid, "data", "uchar"); > c += fwrite (fid, ck_size, "uint32", 0, BYTEORDER); > > if (c < 25) > fclose (fid); > error ("wavread: writing to file failed"); > endif > > ## interleave samples > yi = reshape (y', n*channels, 1); > > ## scale samples > switch (bits_per_sample) > case 8 > yi = round (yi*127.5 + 127.5); > case 16 > yi = floor (yi*32767.5); > case 32 > yi = floor (yi*2147483647.5); > endswitch > > ## write to file > c = fwrite (fid, yi, format, 0, BYTEORDER); > > fclose (fid); > > endfunction
Randy, Please send this report to bug@octave.org as this is a bug report in a core octave function D.