Edinburgh Speech Tools  2.4-release
 All Classes Functions Variables Typedefs Enumerations Enumerator Friends Pages
gen_audio.cc
1 /*************************************************************************/
2 /* */
3 /* Centre for Speech Technology Research */
4 /* University of Edinburgh, UK */
5 /* Copyright (c) 1995,1996 */
6 /* All Rights Reserved. */
7 /* */
8 /* Permission is hereby granted, free of charge, to use and distribute */
9 /* this software and its documentation without restriction, including */
10 /* without limitation the rights to use, copy, modify, merge, publish, */
11 /* distribute, sublicense, and/or sell copies of this work, and to */
12 /* permit persons to whom this work is furnished to do so, subject to */
13 /* the following conditions: */
14 /* 1. The code must retain the above copyright notice, this list of */
15 /* conditions and the following disclaimer. */
16 /* 2. Any modifications must be clearly marked as such. */
17 /* 3. Original authors' names are not deleted. */
18 /* 4. The authors' names are not used to endorse or promote products */
19 /* derived from this software without specific prior written */
20 /* permission. */
21 /* */
22 /* THE UNIVERSITY OF EDINBURGH AND THE CONTRIBUTORS TO THIS WORK */
23 /* DISCLAIM ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING */
24 /* ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO EVENT */
25 /* SHALL THE UNIVERSITY OF EDINBURGH NOR THE CONTRIBUTORS BE LIABLE */
26 /* FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES */
27 /* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN */
28 /* AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, */
29 /* ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF */
30 /* THIS SOFTWARE. */
31 /* */
32 /*************************************************************************/
33 /* Author : Paul Taylor */
34 /* Date : March 95 */
35 /*-----------------------------------------------------------------------*/
36 /* Generalised playback function */
37 /* */
38 /*=======================================================================*/
39 
40 #include <cstdlib>
41 #include <iostream>
42 #include <cstring>
43 #include <cmath>
44 #include <fcntl.h>
45 #include "EST_system.h"
46 #include "EST_socket.h"
47 #include "EST_Option.h"
48 #include "EST_Wave.h"
49 #include "EST_io_aux.h"
50 #include "audioP.h"
51 #include "EST_audio.h"
52 #include "EST_wave_aux.h"
53 
54 static int play_sunau_wave(EST_Wave &inwave, EST_Option &al);
55 static int play_socket_wave(EST_Wave &inwave, EST_Option &al);
56 static int play_aucomm_wave(EST_Wave &inwave, EST_Option &al);
57 
58 static int record_sunau_wave(EST_Wave &wave, EST_Option &al);
59 
60 int play_wave(EST_Wave &inwave, EST_Option &al)
61 {
62  EST_String protocol;
63  EST_Wave wtmp;
64  EST_Wave *toplay;
65  char *quality;
66  char *sr;
67 
68  if ((sr = getenv("NA_PLAY_HOST")) != NULL)
69  if (!al.present("-display"))
70  al.add_item("-display", sr);
71 
72  if ((quality = getenv("NA_PLAY_QUALITY")) != NULL)
73  if (!al.present("-quality"))
74  al.add_item("-quality", quality);
75 
76  if (al.present("-p"))
77  protocol = al.val("-p");
78  else if ((sr=getenv("NA_PLAY_PROTOCOL")) != NULL)
79  protocol = sr;
80  else if (protocol == "")
81  {
82  if (nas_supported)
83  protocol = "netaudio"; // the default protocol
84  else if (esd_supported)
85  protocol = "esdaudio";
86  else if (sun16_supported)
87  protocol = "sun16audio";
88  else if (freebsd16_supported)
89  protocol = "freebsd16audio";
90  else if (linux16_supported)
91  protocol = "linux16audio";
92  else if (irix_supported)
93  protocol = "irixaudio";
94  else if (macosx_supported)
95  protocol = "macosxaudio";
96  else if (win32audio_supported)
97  protocol = "win32audio";
98  else if (mplayer_supported)
99  protocol = "mplayeraudio";
100  else
101  protocol = "sunaudio";
102  }
103 
104  // OS X can handle multichannel audio, don't know about other systems.
105  if (inwave.num_channels() > 1 && upcase(protocol) != "MACOSXAUDIO" )
106  {
107  wave_combine_channels(wtmp,inwave);
108  toplay = &wtmp;
109  }
110  else
111  toplay = &inwave;
112 
113  if (upcase(protocol) == "NETAUDIO")
114  return play_nas_wave(*toplay,al);
115  else if (upcase(protocol) == "ESDAUDIO")
116  return play_esd_wave(*toplay,al);
117  else if (upcase(protocol) == "SUNAUDIO")
118  return play_sunau_wave(*toplay,al);
119  else if (upcase(protocol) == "SUN16AUDIO")
120  return play_sun16_wave(*toplay,al);
121  else if ((upcase(protocol) == "FREEBSD16AUDIO") ||
122  (upcase(protocol) == "LINUX16AUDIO"))
123  return play_linux_wave(*toplay,al);
124  else if (upcase(protocol) == "IRIXAUDIO")
125  return play_irix_wave(*toplay,al);
126  else if (upcase(protocol) == "MACOSXAUDIO")
127  return play_macosx_wave(*toplay,al);
128  else if (upcase(protocol) == "MPLAYERAUDIO")
129  return play_mplayer_wave(*toplay,al);
130  else if (upcase(protocol) == "WIN32AUDIO")
131  return play_win32audio_wave(*toplay,al);
132  else if (upcase(protocol) == "AUDIO_COMMAND")
133  return play_aucomm_wave(*toplay,al);
134  else if (upcase(protocol) == "SOCKET")
135  return play_socket_wave(*toplay,al);
136  else
137  {
138  cerr << "Unknown audio server protocol " << protocol << endl;
139  return -1;
140  }
141 }
142 
143 static int play_socket_wave(EST_Wave &inwave, EST_Option &al)
144 {
145  // Send inwave down the given fd (a socket)
146  SOCKET_FD fd;
147  EST_String otype;
148  EST_String tmpfile = make_tmp_filename();
149 
150  if (al.present("socket_fd"))
151  fd = al.ival("socket_fd");
152  else
153  {
154  cerr << "Socket audio mode: no socket_fd specified" << endl;
155  return -1;
156  }
157 
158  if (al.present("socket_otype"))
159  otype = al.val("socket_otype"); // file type to send to client
160  else
161  otype = "riff";
162 
163  inwave.save(tmpfile,otype);
164 
165  // Because the client may receive many different types of file
166  // I send WV\n to it before the file itself
167  send(fd,"WV\n",3,0);
168  socket_send_file(fd,tmpfile);
169  unlink(tmpfile);
170 
171  return 0;
172 }
173 
174 static int play_aucomm_wave(EST_Wave &inwave, EST_Option &al)
175 {
176  // Play wave by specified command
177  EST_String usrcommand, otype;
178  char tmpfile[2048];
179  char pref[2048];
180 
181  if (al.present("-command"))
182  usrcommand = al.val("-command");
183  else if (getenv("NA_PLAY_COMMAND") != NULL)
184  usrcommand = getenv("NA_PLAY_COMMAND");
185  else
186  {
187  cerr << "Audio protocol set to COMMAND but no command specified\n";
188  return -1;
189  }
190 
191  sprintf(tmpfile,"/tmp/audiofile_%05ld",(long)getpid());
192 
193  if (al.present("-rate"))
194  inwave.resample(al.ival("-rate"));
195  if (al.present("-otype"))
196  otype = al.val("-otype");
197  else
198  otype = "raw";
199 
200  if (inwave.save(tmpfile,otype) != write_ok)
201  {
202  cerr << "Audio writing file \"" << tmpfile << "\" in type \"" <<
203  otype << " failed " << endl;
204  return -1;
205  }
206 
207  sprintf(pref,"FILE=%s;SR=%d;",tmpfile,inwave.sample_rate());
208 
209  system((EST_String)pref+usrcommand.unquote('"'));
210 
211  unlink(tmpfile); // so we don't fill up /tmp
212 
213  return 0;
214 }
215 
216 static int play_sunau_wave(EST_Wave &inwave, EST_Option &al)
217 {
218  // Play wave through /dev/audio using 8K ulaw encoding
219  // works for Suns as well as Linux and FreeBSD machines
220  int rcode;
221  const char *audiodevice;
222 
223  inwave.resample(8000);
224 
225  if (al.present("-audiodevice"))
226  audiodevice = al.val("-audiodevice");
227  else
228  audiodevice = "/dev/audio";
229 
230  // Should really do something cute about checking if /dev/audio
231  // is not in use
232  rcode = inwave.save(audiodevice,"ulaw");
233 
234  return rcode;
235 
236 }
237 
238 EST_String options_supported_audio(void)
239 {
240  // returns list of supported audio types
241  EST_String audios = "";
242 
243  audios += "sunaudio"; // we always support this in spite of the hardware
244 
245  audios += " audio_command";
246  if (nas_supported)
247  audios += " netaudio";
248  else if (esd_supported)
249  audios += " esdaudio";
250  if (sun16_supported)
251  audios += " sun16audio";
252  if (freebsd16_supported)
253  audios += " freebsd16audio";
254  if (linux16_supported)
255  audios += " linux16audio";
256  if (irix_supported)
257  audios += " irixaudio";
258  if (mplayer_supported)
259  audios += " mplayeraudio";
260  if (macosx_supported)
261  audios += "macosxaudio";
262  if (win32audio_supported)
263  audios += " win32audio";
264  if (os2audio_supported)
265  audios += " os2audio";
266 
267  return audios;
268 }
269 
270 int record_wave(EST_Wave &wave, EST_Option &al)
271 {
272  // Record wave from audio device
273  char *sr;
274  EST_String protocol;
275 
276  // For archaic reasons, if you are using NAS use DISPLAY or
277  // AUDIOSERVER
278  if ((sr = getenv("NA_PLAY_HOST")) != NULL)
279  if (!al.present("-display"))
280  al.add_item("-display", sr);
281 
282  if (al.present("-p"))
283  protocol = al.val("-p");
284  else if ((sr=getenv("NA_PLAY_PROTOCOL")) != NULL)
285  protocol = sr;
286  else if (protocol == "")
287  {
288  if (nas_supported)
289  protocol = "netaudio"; // the default protocol
290  else if (esd_supported)
291  protocol = "esdaudio"; // the default protocol
292  else if (sun16_supported)
293  protocol = "sun16audio";
294  else if (freebsd16_supported)
295  protocol = "freebsd16audio";
296  else if (linux16_supported)
297  protocol = "linux16audio";
298  else if (irix_supported)
299  protocol = "irixaudio";
300  else if (win32audio_supported)
301  protocol = "win32audio";
302  else if (mplayer_supported)
303  protocol = "mplayeraudio";
304  else
305  protocol = "sunaudio";
306  }
307 
308  if (upcase(protocol) == "NETAUDIO")
309  return record_nas_wave(wave,al);
310  else if (upcase(protocol) == "ESDAUDIO")
311  return record_esd_wave(wave,al);
312  else if (upcase(protocol) == "SUN16AUDIO")
313  return record_sun16_wave(wave,al);
314  else if ((upcase(protocol) == "FREEBSD16AUDIO") ||
315  (upcase(protocol) == "LINUX16AUDIO"))
316  return record_linux_wave(wave,al);
317  else if (upcase(protocol) == "SUNAUDIO")
318  return record_sunau_wave(wave,al);
319  else
320  {
321  cerr << "NA_RECORD: \"" << protocol <<
322  "\" EST current has no record support" << endl;
323  return -1;
324  }
325 }
326 
327 static int record_sunau_wave(EST_Wave &wave, EST_Option &al)
328 {
329  int num_samples,i,r,n;
330  int audio;
331  unsigned char *ulawwave;
332  short *waveform;
333  const int AUDIOBUFFSIZE = 256;
334  const char *audiodevice;
335 
336  if (al.present("-audiodevice"))
337  audiodevice = al.val("-audiodevice");
338  else
339  audiodevice = "/dev/audio";
340 
341  if ((audio = open(audiodevice, O_RDONLY)) == -1)
342  {
343  cerr << "SUN16: can't open " << audiodevice << " for reading" << endl;
344  return -1;
345  }
346 
347  num_samples = (int)(8000*al.fval("-time"));
348  ulawwave = walloc(unsigned char,num_samples);
349 
350  for (r=i=0; i < num_samples; i+= r)
351  {
352  if (num_samples > i+AUDIOBUFFSIZE)
353  n = AUDIOBUFFSIZE;
354  else
355  n = num_samples-i;
356  r = read(audio,&ulawwave[i], n);
357  if (r <= 0)
358  {
359  cerr << "sunaudio: failed to read from audio device" << endl;
360  close(audio);
361  wfree(ulawwave);
362  return -1;
363  }
364  }
365 
366  wave.resize(num_samples);
367  wave.set_sample_rate(8000);
368  waveform = wave.values().memory();
369 
370  ulaw_to_short(ulawwave,waveform,num_samples);
371  wave.resample(al.ival("-sample_rate"));
372 
373  close(audio);
374  wfree(ulawwave);
375  return 0;
376 }
377