Description: Allow alaw output (asterisk interoperability)
Forwarded: https://github.com/festvox/speech_tools/pull/13

--- a/include/EST_wave_aux.h
+++ b/include/EST_wave_aux.h
@@ -97,6 +97,7 @@
 void uchar_to_short(const unsigned char *chars,short *data,int length);
 void short_to_char(const short *data,unsigned char *chars,int length);
 void short_to_ulaw(const short *data,unsigned char *ulaw,int length);
+void short_to_alaw(const short *data,unsigned char *alaw,int length);
 
 // Used when setting Waves in Features
 VAL_REGISTER_CLASS_DCLS(wave,EST_Wave)
--- a/speech_class/EST_wave_aux.cc
+++ b/speech_class/EST_wave_aux.cc
@@ -287,7 +287,7 @@
 	"    endian)\n\n"
 	"-iswap  Swap bytes. (For use on an unheadered input file)\n\n"
 	"-istype <string> Sample type in an unheadered input file:\n"
-	"     short, mulaw, byte, ascii\n\n"
+	"     short, alaw, mulaw, byte, ascii\n\n"
 	"-c <string>  Select a single channel (starts from 0). \n"
 	"    Waveforms can have multiple channels. This option \n"
         "    extracts a single channel for progcessing and \n"
@@ -318,7 +318,7 @@
 	"    Intel, Alpha, DEC Mips, Vax are LSB \n"
 	"    (little endian)\n\n"
 	"-oswap Swap bytes when saving to output\n\n"+
-	"-ostype <string> Output sample type: short, mulaw, byte or ascii\n\n";
+	"-ostype <string> Output sample type: short, alaw, mulaw, byte or ascii\n\n";
 }
 
 Declare_TNamedEnum(EST_sample_type_t)
--- a/speech_class/EST_WaveFile.cc
+++ b/speech_class/EST_WaveFile.cc
@@ -312,6 +312,27 @@
     return save_using(save_wave_ulaw, fp, localwv, stype, bo);
 }
 
+EST_read_status EST_WaveFile::load_alaw(EST_TokenStream &ts,
+  EST_Wave &wv,
+  int rate,
+  EST_sample_type_t stype, int bo, int nchan,
+  int offset, int length)
+{
+  return load_using(load_wave_alaw,
+    ts, wv, rate, 
+    stype, bo, nchan,
+    offset, length);
+}
+
+EST_write_status EST_WaveFile::save_alaw(FILE *fp,
+ const EST_Wave &wv,
+ EST_sample_type_t stype, int bo)
+{
+    EST_Wave localwv = wv;
+    localwv.resample(8000);
+    return save_using(save_wave_alaw, fp, localwv, stype, bo);
+}
+
 static int parse_esps_r_option(EST_String arg, int &offset, int &length)
 {
     EST_String s, e;
@@ -382,6 +403,11 @@
 	al.add_item("-itype","ulaw");
 	al.add_item("-f","8000");
     }
+    if (al.present("-alaw"))
+    {
+al.add_item("-itype","alaw");
+al.add_item("-f","8000");
+    }
     if (al.present("-iswap"))
 	al.add_item("-ibo","other");
 
@@ -452,6 +478,11 @@
 	cerr << "Cannot recognize file format or cannot access file: \"" << in_file << "\"\n";
 	return read_error;
     }
+    if (file_type == "alaw")
+    {
+sample_rate = 8000;
+sample_type = "alaw";
+    }
 
     if (al.present("-start") || al.present("-end") 
 	|| al.present("-to") || al.present("-from"))
--- a/speech_class/EST_WaveFile.h
+++ b/speech_class/EST_WaveFile.h
@@ -57,7 +57,8 @@
   wff_aiff,
   wff_riff,
   wff_raw,
-  wff_ulaw
+  wff_ulaw,
+  wff_alaw
 } EST_WaveFileType;
 
 class EST_WaveFile {
@@ -125,6 +126,9 @@
   static EST_write_status save_ulaw(SaveWave_TokenStreamArgs);
   static EST_read_status load_ulaw(LoadWave_TokenStreamArgs);
 
+  static EST_write_status save_alaw(SaveWave_TokenStreamArgs);
+  static EST_read_status load_alaw(LoadWave_TokenStreamArgs);
+
   static EST_TNamedEnumI<EST_WaveFileType, Info> map;
 
   static EST_String options_supported(void);
--- a/speech_class/EST_wave_io.cc
+++ b/speech_class/EST_wave_io.cc
@@ -107,15 +107,17 @@
 {
     const char *c;
     switch (sample_type) {
-    case st_unknown:  
+    case st_unknown:
 	c = ""; break;
-    case st_schar:  
+    case st_schar:
 	c = "PCM-1"; break;
+    case st_alaw:
+	c = "ALAW"; break;
     case st_mulaw:
 	c = "ULAW"; break;
-    case st_short: 
+    case st_short:
 	c = "pcm"; break;
-    case st_int:   
+    case st_int:
 	c = "PCM-4"; break;
     case st_float:
 	c = "REAL"; break;
@@ -141,6 +143,9 @@
 	     (EST_strcasecmp(type,"mu-law",NULL) == 0) ||
 	     (EST_strcasecmp(type,"mulaw",NULL) == 0))
 	return st_mulaw;
+    else if ((EST_strcasecmp(type,"ALAW",NULL) == 0) ||
+	     (EST_strcasecmp(type,"A-LAW",NULL) == 0))
+	return st_alaw;
     else if (strcmp(type,"alaw") == 0)
 	return st_alaw;
     else if (strcmp(type,"PCM-1") == 0)
@@ -192,6 +197,11 @@
 	byte_order = wstrdup((EST_BIG_ENDIAN ? "10" : "01"));
 	sample_coding = wstrdup("ULAW");
     }
+    if (streq(byte_order,"a-law"))
+    {
+	byte_order = wstrdup((EST_BIG_ENDIAN ? "10" : "01"));
+	sample_coding = wstrdup("ALAW");
+    }
 
     /* code for reading in Tony Robinson's shorten files.
        This is a temporary fix which calls the unshorten program on the
@@ -484,12 +494,13 @@
 	/* The follow are registered proprietary WAVE formats  (?) */
     case WAVE_FORMAT_MULAW:
 	actual_sample_type = st_mulaw; break;
+    case WAVE_FORMAT_ALAW:
+	actual_sample_type = st_alaw; break;
     case WAVE_FORMAT_ADPCM:
-	fprintf(stderr, "RIFF file: unsupported proprietary sample format ADPCM\n"); 
+	fprintf(stderr, "RIFF file: unsupported proprietary sample format ADPCM\n");
 	actual_sample_type = st_short;
 	break;
 	/*	  actual_sample_type = st_adpcm; break; */ /* yes but which adpcm ! */
-    case WAVE_FORMAT_ALAW:
     default:
 	fprintf(stderr, "RIFF file: unknown sample format\n");
 	actual_sample_type = st_short;
@@ -854,50 +865,103 @@
 {
     unsigned char *ulaw;
     int data_length,samps;
-    
+
     ts.seek_end();
     samps = ts.tell();
-    
+
     if (length == 0)
 	data_length = samps - offset;
     else
 	data_length = length;
-    
+
     ulaw = walloc(unsigned char, data_length);
     ts.seek(offset);
     if (ts.fread(ulaw,1,data_length) != data_length)
     {
-	wfree(ulaw); 
+	wfree(ulaw);
 	return misc_read_error;
     }
-    
+
     *data = walloc(short,data_length);
     ulaw_to_short(ulaw,*data,data_length);
     wfree(ulaw);
-    
+
     *num_samples = data_length;
     *sample_rate = 8000;
     *num_channels = 1;
     *sample_type = st_short;
     *word_size = 2;
     *bo = EST_NATIVE_BO;
-    
+
     return format_ok;
 }
 
 enum EST_write_status save_wave_ulaw(FILE *fp, const short *data, int offset,
-				     int num_samples, int num_channels, 
-				     int sample_rate, 
+				     int num_samples, int num_channels,
+				     int sample_rate,
 				     enum EST_sample_type_t sample_type, int bo)
 {
     (void)sample_rate;
     (void)sample_type;
     return save_wave_raw(fp,data,offset,num_samples,num_channels,
 			 8000,st_mulaw,bo);
-    
-    
+
+
 }
 
+enum EST_read_status load_wave_alaw(EST_TokenStream &ts, short **data, int
+				    *num_samples, int *num_channels, int *word_size, int
+				    *sample_rate, enum EST_sample_type_t *sample_type, int *bo,
+				    int offset, int length)
+
+{
+    unsigned char *alaw;
+    int data_length,samps;
+
+    ts.seek_end();
+    samps = ts.tell();
+
+    if (length == 0)
+	data_length = samps - offset;
+    else
+	data_length = length;
+
+    alaw = walloc(unsigned char, data_length);
+    ts.seek(offset);
+    if (ts.fread(alaw,1,data_length) != data_length)
+    {
+	wfree(alaw);
+	return misc_read_error;
+    }
+
+    *data = walloc(short,data_length);
+    alaw_to_short(alaw,*data,data_length);
+    wfree(alaw);
+
+    *num_samples = data_length;
+    *sample_rate = 8000;
+    *num_channels = 1;
+    *sample_type = st_short;
+    *word_size = 2;
+    *bo = EST_NATIVE_BO;
+
+    return format_ok;
+}
+
+enum EST_write_status save_wave_alaw(FILE *fp, const short *data, int offset,
+				     int num_samples, int num_channels,
+				     int sample_rate,
+				     enum EST_sample_type_t sample_type, int bo)
+{
+    (void)sample_rate;
+    (void)sample_type;
+    return save_wave_raw(fp,data,offset,num_samples,num_channels,
+			 8000,st_alaw,bo);
+
+
+}
+
+
 /*=======================================================================*/
 /* Sun and Next snd files                                                */
 /*=======================================================================*/
--- a/speech_class/EST_wave_utils.cc
+++ b/speech_class/EST_wave_utils.cc
@@ -53,7 +53,9 @@
 #include "EST_error.h"
 
 static short st_ulaw_to_short(unsigned char ulawbyte);
+static short st_alaw_to_short(unsigned char alawbyte);
 static unsigned char st_short_to_ulaw(short sample);
+static unsigned char st_short_to_alaw(short sample);
 
 /*
  * This table is 
@@ -202,6 +204,16 @@
     
 }
 
+void short_to_alaw(const short *data,unsigned char *alaw,int length)
+{
+    /* Convert alaw to shorts */
+    int i;
+
+    for (i=0; i<length; i++)
+	alaw[i] = st_short_to_alaw(data[i]);  /* alaw convert */
+    
+}
+
 short *convert_raw_data(unsigned char *file_data,int data_length,
 			enum EST_sample_type_t sample_type,int bo)
 {
@@ -230,6 +242,13 @@
 	wfree(file_data);
 	return d;
     }
+    else if (sample_type == st_alaw)
+    {
+	d = walloc(short,data_length);
+	alaw_to_short(file_data,d,data_length);
+	wfree(file_data);
+	return d;
+    }
 #if 0
     else if (sample_type == st_adpcm)
     {   /* Not really checked yet */
@@ -279,6 +298,16 @@
 	if (n != (num_channels * num_samples))
 	    return misc_write_error;
     }
+    else if (sample_type == st_alaw)
+    {
+	unsigned char *alaw = walloc(unsigned char,num_samples*num_channels);
+	short_to_alaw(data+(offset*num_channels),
+		      alaw,num_samples*num_channels);
+	n = fwrite(alaw,1,num_channels * num_samples,fp);	
+	wfree(alaw);
+	if (n != (num_channels * num_samples))
+	    return misc_write_error;
+    }
     else if (sample_type == st_ascii)
     {
 	for (i=offset*num_channels; i < num_samples*num_channels; i++)
@@ -354,6 +383,7 @@
       case st_schar:  
 	word_size = 1;	break;
       case st_mulaw:
+      case st_alaw:
 	word_size = 1;	break;
 #if 0
       case st_adpcm:  /* maybe I mean 0.5 */
@@ -420,6 +450,7 @@
       case st_short:   return "short";
       case st_shorten:   return "shorten";
       case st_mulaw:   return "ulaw";
+      case st_alaw:    return "alaw";
       case st_schar:    return "char";
       case st_uchar:    return "unsignedchar";
       case st_int:     return "int";
@@ -531,6 +562,77 @@
     return sample;
 }
 
+
+/* The following is copied from sox:g711.c */
+/*
+ * This source code is a product of Sun Microsystems, Inc. and is provided
+ * for unrestricted use.  Users may copy or modify this source code without
+ * charge.
+ * [ no warranties or liabilities whatsoever; see there for details. ]
+ */
+/* copy from CCITT G.711 specifications */
+static unsigned char _u2a[128] = {	/* u- to A-law conversions */
+	1,	1,	2,	2,	3,	3,	4,	4,
+	5,	5,	6,	6,	7,	7,	8,	8,
+	9,	10,	11,	12,	13,	14,	15,	16,
+	17,	18,	19,	20,	21,	22,	23,	24,
+	25,	27,	29,	31,	33,	34,	35,	36,
+	37,	38,	39,	40,	41,	42,	43,	44,
+	46,	48,	49,	50,	51,	52,	53,	54,
+	55,	56,	57,	58,	59,	60,	61,	62,
+	64,	65,	66,	67,	68,	69,	70,	71,
+	72,	73,	74,	75,	76,	77,	78,	79,
+	80,	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};
+
+static unsigned char _a2u[128] = {		/* A- to u-law conversions */
+	1,	3,	5,	7,	9,	11,	13,	15,
+	16,	17,	18,	19,	20,	21,	22,	23,
+	24,	25,	26,	27,	28,	29,	30,	31,
+	32,	32,	33,	33,	34,	34,	35,	35,
+	36,	37,	38,	39,	40,	41,	42,	43,
+	44,	45,	46,	47,	48,	48,	49,	49,
+	50,	51,	52,	53,	54,	55,	56,	57,
+	58,	59,	60,	61,	62,	63,	64,	64,
+	65,	66,	67,	68,	69,	70,	71,	72,
+	73,	74,	75,	76,	77,	78,	79,	80,
+	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};
+
+/* A-law to u-law conversion */
+static inline unsigned char st_alaw2ulaw(
+	unsigned char	aval)
+{
+	aval &= 0xff;
+	return (unsigned char) ((aval & 0x80) ? (0xFF ^ _a2u[aval ^ 0xD5]) :
+	    (0x7F ^ _a2u[aval ^ 0x55]));
+}
+
+/* u-law to A-law conversion */
+static inline unsigned char st_ulaw2alaw(
+	unsigned char	uval)
+{
+	uval &= 0xff;
+	return (unsigned char) ((uval & 0x80) ? (0xD5 ^ (_u2a[0xFF ^ uval] - 1)) :
+	    (unsigned char) (0x55 ^ (_u2a[0x7F ^ uval] - 1)));
+}
+/* end of Sun code */
+
+/* This is somewhat simple-minded, but ... */
+static unsigned char st_short_to_alaw(short sample)
+{
+    return st_ulaw2alaw(st_short_to_ulaw(sample));
+}
+
+
 /*
  * C O N V E R T   T O   I E E E   E X T E N D E D
  */
--- a/speech_class/waveP.h
+++ b/speech_class/waveP.h
@@ -41,7 +41,7 @@
 
 #include <stdio.h>
 
-/* The follow two (raw and ulaw) cannot be in the table as they cannot */
+/* The follow two (raw, alaw and ulaw) cannot be in the table as they cannot */
 /* identify themselves from files (both are unheadered)                */
 enum EST_read_status load_wave_raw(EST_TokenStream &ts, short **data, int
 	 *num_samples, int *num_channels, int *word_size, int
@@ -61,6 +61,15 @@
 				int length, int num_channels, 
 				int sample_rate,
 				enum EST_sample_type_t, int bo);
+
+enum EST_read_status load_wave_alaw(EST_TokenStream &ts, short **data, int
+	 *num_samples, int *num_channels, int *word_size, int
+	 *sample_rate,  enum EST_sample_type_t *sample_type, int *bo, int
+	 offset, int length);
+enum EST_write_status save_wave_alaw(FILE *fp, const short *data, int offset,
+				int length, int num_channels, 
+				int sample_rate,
+				enum EST_sample_type_t, int bo);
 
 enum EST_read_status load_wave_nist(EST_TokenStream &ts, short **data, int
 	 *num_samples, int *num_channels, int *word_size, int
--- a/testsuite/correct/ch_wave_script.out
+++ b/testsuite/correct/ch_wave_script.out
@@ -77,7 +77,7 @@
 -iswap  Swap bytes. (For use on an unheadered input file)
 
 -istype <string> Sample type in an unheadered input file:
-     short, mulaw, byte, ascii
+     short, alaw, mulaw, byte, ascii
 
 -c <string>  Select a single channel (starts from 0). 
     Waveforms can have multiple channels. This option 
@@ -112,7 +112,7 @@
 
 -oswap Swap bytes when saving to output
 
--ostype <string> Output sample type: short, mulaw, byte or ascii
+-ostype <string> Output sample type: short, alaw, mulaw, byte or ascii
 
 -scale <float> Scaling factor. Increase or descrease the amplitude
     of the whole waveform by the factor given