function ft_realtime_pooraudioproxy(cfg)

% FT_REALTIME_POORAUDIOPROXY reads continuous data from the sound card using the

% standard Matlab API and writes it to a FieldTrip buffer. This proxy has poor timing

% and will produce dropped audio frames between blocks. Also the Matlab documentation

% warns about using this API for long recordings because this will fill up memory and

% degrade performance.

%

% The FieldTrip buffer is a network transparent server that allows the acquisition

% client to stream data to it. An analysis client can connect to read the data upon

% request. Multiple clients can connect simultaneously, each analyzing a specific

% aspect of the data concurrently.

%

% Use as

% ft_realtime_pooraudioproxy(cfg)

%

% The audio-specific configuration structure can contain

% cfg.channel = number of channels (1 or 2, default=2)

% cfg.blocksize = size of recorded audio blocks in seconds (default=1)

% cfg.fsample = audio sampling frequency in Hz (default = 44100)

% cfg.nbits = recording depth in bits (default = 16)

%

% Note that currently, the sound will be buffered in double precision irrespective of the sampling bit depth.

%

% The target to write the data to is configured as

% cfg.target.datafile = string, target destination for the data (default = 'buffer://localhost:1972')

% cfg.target.dataformat = string, default is determined automatic

%

% Finally, there is an option for showing debug output

% cfg.debug = show sample time and clock time (default = 'yes')

%

% To stop this realtime function, you have to press Ctrl-C

%

% See also FT_REALTIME_SIGNALPROXY, FT_REALTIME_SIGNALVIEWER

% Copyright (C) 2010, Stefan Klanke & Robert Oostenveld

%

% This file is part of FieldTrip, see http://www.fieldtriptoolbox.org

% for the documentation and details.

%

% FieldTrip 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 3 of the License, or

% (at your option) any later version.

%

% FieldTrip 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 FieldTrip. If not, see <http://www.gnu.org/licenses/>.

%

% $Id$

% set the defaults

if ~isfield(cfg, 'target'), cfg.target = []; end

if ~isfield(cfg, 'blocksize'), cfg.blocksize = 1; end % in seconds

if ~isfield(cfg, 'channel'), cfg.channel = 2; end % default is stereo

if ~isfield(cfg, 'fsample'), cfg.fsample = 44100; end % in Hz

if ~isfield(cfg, 'nbits'), cfg.nbits = 16; end % default 16 bit

if ~isfield(cfg, 'debug'), cfg.debug = 'yes'; end

if ~isfield(cfg.target, 'datafile'), cfg.target.datafile = 'buffer://localhost:1972'; end

if ~isfield(cfg.target, 'dataformat'), cfg.target.dataformat = []; end % default is to use autodetection of the output format

% it seems that audiorecorder can only record integer multiples of 0.5 seconds

% this was observed on OSX with Matlab 2008a, but has not been confirmed on

% other platforms

newblocksize = round(cfg.blocksize*2)/2; % round to nearest half or full second

if newblocksize<0.5

newblocksize = 0.5;

end

if newblocksize~=cfg.blocksize

ft_warning('sestting cfg.blocksize to %f', newblocksize);

cfg.blocksize = newblocksize;

end

hdr = [];

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% create a FieldTrip compatible header structure

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

hdr.Fs = cfg.fsample;

hdr.nChans = cfg.channel;

hdr.nSamples = 0;

hdr.nSamplesPre = 0;

hdr.nTrials = 1;

if cfg.channel == 2

hdr.label = {'Audio Left','Audio Right'};

elseif cfg.channel == 1

hdr.label = {'Audio Left'};

else

error 'Invalid channel number (only 1 or 2 are allowed)';

end

hdr.FirstTimeStamp = nan;

hdr.TimeStampPerSample = nan;

% create an audio recorder object

REC = audiorecorder(cfg.fsample, cfg.nbits, cfg.channel);

count = 0;

endsample = 0;

stopwatch = tic;

while true

% tell the audio recorder to read (and wait for) cfg.blocksize samples, and then get the data into the matrix x

recordblocking(REC, cfg.blocksize);

x = getaudiodata(REC);

endsample = endsample + size(x,1);

% since this is a "poor" audio proxy, the clock time will move faster than the sample time here

% in general, those two should be the same (or have a small constant offset)

if strcmp(cfg.debug, 'yes')

fprintf('number of samples acquired = %i, sample time = %f, clock time = %f\n', endsample, endsample/hdr.Fs, toc(stopwatch));

end

% increase the internal counter of blocks, so we know whether to append the new data or not

count = count+1;

if count==1

% flush the file, write the header and subsequently write the data segment

ft_write_data(cfg.target.datafile, x', 'header', hdr, 'dataformat', cfg.target.dataformat, 'append', false);

else

% write the data segment

ft_write_data(cfg.target.datafile, x', 'append', true);

end % if count==1

hdr.nSamples = endsample;

end % while again