[1.30] Add option for a pause between words: command option -g or espeak_SetParameter(espeak_WORDGAP)

Add different intonation styles (attribute "intonation"  in voices files).


git-svn-id: https://espeak.svn.sourceforge.net/svnroot/espeak/trunk@114 d46cf337-b52f-0410-862d-fd96e6ae7743
This commit is contained in:
jonsd
2007-12-31 15:56:11 +00:00
parent fb7230a912
commit dc09dd11f1
58 changed files with 684 additions and 497 deletions
+1 -1
View File
@@ -35,7 +35,7 @@ Installation and Use
3. Example commands:
*help speak
should now show version 3.10
should now show version 3.30 (or whatever is the latest version)
*speak -h
shows the available commands
+101 -197
View File
@@ -1,23 +1,22 @@
/**************************************************************************
* Copyright (C) 2005,2006 by Jonathan Duddington *
* jonsd@users.sourceforge.net *
* *
* This program 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 of the License, or *
* (at your option) any later version. *
* *
* This program 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 this program; if not, write to the *
* Free Software Foundation, Inc., *
* 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. *
***************************************************************************/
/***************************************************************************
* Copyright (C) 2005 to 2007 by Jonathan Duddington *
* email: jonsd@users.sourceforge.net *
* *
* This program 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. *
* *
* This program 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 this program; if not, write see: *
* <http://www.gnu.org/licenses/>. *
***************************************************************************/
#define USE_MODULE
@@ -30,9 +29,9 @@
#include "speech.h"
#include "speak_lib.h"
#include "voice.h"
#include "phoneme.h"
#include "synthesize.h"
#include "voice.h"
#include "translate.h"
#define os_X 0x20000
@@ -65,10 +64,11 @@ extern int current_source_index;
FILE *f_text;
int (* uri_callback)(int, const char *, const char *) = NULL;
int (* phoneme_callback)(const char *) = NULL;
int amp = 8; // default
char path_home[80] = "";
char path_home[N_PATH_HOME] = "";
char wavefile[120];
char textbuffile[L_tmpnam];
int sample_rate_index; // current value
@@ -81,7 +81,7 @@ static void *module_data;
static int callback_inhibit = 0;
static int more_text=0;
#define N_VOICE_NAMES 40
#define N_VOICE_NAMES 60
static char *voice_names[40];
#define N_STATIC_BUF 8000
@@ -91,17 +91,17 @@ static _kernel_oserror errblk;
USHORT voice_pcnt[N_PEAKS+1][3];
voice_t voice_data;
voice_t *voice;
static const char *help_text =
"\nspeak [options] [\"<words>\"]\n\n"
"-f <text file> Text file to speak\n"
"--stdin Read text input from stdin instead of a file\n\n"
//"--stdin Read text input from stdin instead of a file\n\n"
"If neither -f nor --stdin, <words> are spoken, or if none then text is\n"
"spoken from stdin, each line separately.\n\n"
"-a <integer>\n"
"\t Amplitude, 0 to 200, default is 100\n"
"-g <integer>\n"
"\t Word gap. Pause between words, units of 10mS at the default speed\n"
"-l <integer>\n"
"\t Line length. If not zero (which is the default), consider\n"
"\t lines less than this length as end-of-clause\n"
@@ -118,132 +118,19 @@ static const char *help_text =
"-q\t Quiet, don't produce any speech (may be useful with -x)\n"
"-x\t Write phoneme mnemonics to stdout\n"
"-X\t Write phonemes mnemonics and translation trace to stdout\n"
"--stdout Write speech output to stdout\n"
//"--stdout Write speech output to stdout\n"
"--compile=<voice name>\n"
"\t Compile the pronunciation rules and dictionary in the current\n"
"\t directory. =<voice name> is optional and specifies which language\n"
"--punct=\"<characters>\"\n"
"\t Speak the names of punctuation characters during speaking. If\n"
"\t =<characters> is omitted, all punctuation is spoken.\n"
"--voices=<langauge>\n"
"\t List the available voices for the specified language.\n"
"\t If <language> is omitted, then list all voices.\n";
// additional Latin characters beyond the Latin1 character set
#define MAX_WALPHA 0x233
// indexed by character - 0x100
// 0=not alphabetic, 0xff=lower case, other=value to add to upper case to convert to lower case
static unsigned char walpha_tab[MAX_WALPHA-0xff] = {
1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, // 100
1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, // 110
1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, // 120
0xff,0xff, 1,0xff, 1,0xff, 1,0xff,0xff, 1,0xff, 1,0xff, 1,0xff, 1, // 130
0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff,0xff, 1,0xff, 1,0xff, 1,0xff, // 140
1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, // 150
1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, // 160
1,0xff, 1,0xff, 1,0xff, 1,0xff,0xff, 1,0xff, 1,0xff, 1,0xff,0xff, // 170
0xff, 210, 1,0xff, 1,0xff, 206, 1,0xff, 205, 205, 1,0xff,0xff, 79, 202, // 180
203, 1,0xff, 205, 207,0xff, 211, 209, 1,0xff,0xff,0xff, 211, 213,0xff, 214, // 190
1,0xff, 1,0xff, 1,0xff, 218, 1,0xff, 218,0xff,0xff, 1,0xff, 218, 1, // 1a0
0xff, 217, 217, 1,0xff, 1,0xff, 219, 1,0xff,0xff,0xff, 1,0xff,0xff,0xff, // 1b0
0xff,0xff,0xff,0xff, 2, 1,0xff, 2, 1,0xff, 2, 1,0xff, 1,0xff, 1, // 1c0
0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff,0xff, 1,0xff, // 1d0
1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, // 1e0
0xff, 2, 1,0xff, 1,0xff,0xff,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, // 1f0
1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, // 200
1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, // 210
0xff, 0, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, 1,0xff, // 220
1,0xff, 1,0xff }; // 230
// use ctype.h functions for Latin1 (character < 0x100)
int iswalpha(int c)
{
if(c < 0x100)
return(isalpha(c));
if(c > MAX_WALPHA)
return(0);
return(walpha_tab[c-0x100]);
}
int iswdigit(int c)
{
if(c < 0x100)
return(isdigit(c));
return(0);
}
int iswalnum(int c)
{
if(iswdigit(c))
return(1);
return(iswalpha(c));
}
int towlower(int c)
{
int x;
if(c < 0x100)
return(tolower(c));
if((c > MAX_WALPHA) || ((x = walpha_tab[c-0x100])==0xff))
return(c); // already lower case
return(c + x); // convert to lower case
}
int iswupper(int c)
{
int x;
if(c < 0x100)
return(isupper(c));
if(((c > MAX_WALPHA) || (x = walpha_tab[c-0x100])==0) || (x == 0xff))
return(0);
return(1);
}
int iswlower(int c)
{
if(c < 0x100)
return(islower(c));
if((c > MAX_WALPHA) || (walpha_tab[c-0x100] != 0xff))
return(0);
return(1);
}
int iswspace(int c)
{
if(c < 0x100)
return(isspace(c));
return(0);
}
int iswpunct(int c)
{
if(c < 0x100)
return(ispunct(c));
return(0);
}
const wchar_t *wcschr(const wchar_t *str, int c)
{
while(*str != 0)
{
if(*str == c)
return(str);
str++;
}
return(NULL);
}
const int wcslen(const wchar_t *str)
{
int ix=0;
while(*str != 0)
{
ix++;
}
return(ix);
}
//"--voices=<langauge>\n"
//"\t List the available voices for the specified language.\n"
//"\t If <language> is omitted, then list all voices.\n"
"-k <integer>\n"
"\t Indicate capital letters with: 1=sound, 2=the word \"capitals\",\n"
"\t higher values = a pitch increase (try -k20).\n";
int GetFileLength(const char *filename)
@@ -285,7 +172,7 @@ void ReadVoiceNames2(char *directory)
_kernel_swi_regs regs;
_kernel_oserror *error;
char buf[80];
char directory2[80];
char directory2[sizeof(path_home)+100];
regs.r[0] = 10;
regs.r[1] = (int)directory;
@@ -325,7 +212,7 @@ void ReadVoiceNames2(char *directory)
void ReadVoiceNames()
{//===================
char directory[80];
char directory[sizeof(path_home)+10];
for(n_voice_files=0; n_voice_files<N_VOICE_NAMES; n_voice_files++)
voice_names[n_voice_files] = NULL;
@@ -353,7 +240,7 @@ _kernel_swi_regs regs;
void Free(void *ptr)
/******************/
/*******************/
{ // version of free() for use in RISC_OS module
_kernel_swi_regs regs;
@@ -361,7 +248,7 @@ _kernel_swi_regs regs;
return;
regs.r[0] = 7;
regs.r[2] = (int)ptr;
regs.r[2] = (int)(ptr);
_kernel_swi(0x1e, &regs, &regs); /* OS_Module 7 free memory */
} /* end of Free */
@@ -375,9 +262,12 @@ char *Alloc(int size)
return(p);
}
void Free(void *ptr)
void Free(void **ptr)
{//=================
free(ptr);
if(ptr != NULL)
{
free(ptr);
}
}
#endif
@@ -424,6 +314,7 @@ int initialise(void)
WavegenInit(22050,0);
LoadPhData();
SetVoiceStack(NULL);
SynthesizeInit();
return(0);
}
@@ -447,7 +338,7 @@ void speak_text_string(char *data, int terminator, int len, int wait, int voice_
if((voice_num >= 0) && (voice_num != current_voice_num) && (voice_num < N_VOICE_NAMES))
{
LoadVoice(voice_names[voice_num],0);
SetVoiceByName(voice_names[voice_num]);
WavegenSetVoice(voice);
}
current_voice_num = voice_num;
@@ -613,7 +504,7 @@ void jsd_swi_functions(int *r)
case 6: /* update voice data, r1 = voice_number */
if(r[1] < N_VOICE_NAMES)
{
LoadVoice(voice_names[r[1]],0);
SetVoiceByName(voice_names[r[1]]);
current_voice_num = r[1];
WavegenSetVoice(voice);
}
@@ -698,7 +589,7 @@ _kernel_oserror *swi_handler(int swi_no, int *r, void *pw)
break;
case 9: /* word_gap */
// not implemented
SetParameter(espeakWORDGAP,value,0);
break;
case 10: /* pitch_range */
@@ -723,8 +614,8 @@ _kernel_oserror *swi_handler(int swi_no, int *r, void *pw)
void PitchAdjust(int pitch_adjustment)
{//===================================
int ix, factor;
voice_t *voice = &voice_data;
extern unsigned char pitch_adjust_tab[100];
extern unsigned char pitch_adjust_tab[MAX_PITCH_VALUE+1];
voice->pitch_base = (voice->pitch_base * pitch_adjust_tab[pitch_adjustment])/128;
@@ -776,6 +667,7 @@ void command_line(char *arg_string, int wait)
int value;
int speed;
int amp;
int wordgap;
int speaking = 0;
int flag_stdin = 0;
int flag_compile = 0;
@@ -815,51 +707,19 @@ void command_line(char *arg_string, int wait)
p++;
switch(*p++)
{
case 'b':
option_multibyte = espeakCHARS_8BIT;
break;
case 'h':
printf("\n");
printf("speak text-to-speech: %s\n%s",version_string,help_text);
printf("\nspeak text-to-speech: %s\n%s",version_string,help_text);
return;
case 'a':
amp = param_number(&p);
SetParameter(espeakVOLUME,amp,0);
break;
case 'f':
strcpy(filename,param_string(&p));
break;
case 'k':
option_capitals = param_number(&p);
SetParameter(espeakCAPITALS,option_capitals,0);
break;
case 'l':
option_linelength = param_number(&p);
break;
case 'p':
pitch_adjustment = param_number(&p);
break;
case 'q':
option_quiet = 1;
break;
case 's':
speed = param_number(&p);
SetParameter(espeakRATE,speed,0);
break;
case 'v':
strcpy(voicename,param_string(&p));
break;
case 'w':
option_waveout=1;
strcpy(wavefile,param_string(&p));
break;
case 'x':
option_phonemes = 1;
break;
@@ -868,8 +728,52 @@ void command_line(char *arg_string, int wait)
option_phonemes = 2;
break;
case 'm':
option_ssml = 1;
break;
case 'p':
pitch_adjustment = param_number(&p);
break;
case 'q':
option_quiet = 1;
break;
case 'f':
strncpy0(filename,param_string(&p),sizeof(filename));
break;
case 'l':
option_linelength = param_number(&p);
break;
case 'a':
amp = param_number(&p);
SetParameter(espeakVOLUME,amp,0);
break;
case 's':
speed = param_number(&p);
SetParameter(espeakRATE,speed,0);
break;
case 'g':
wordgap = param_number(&p);
SetParameter(espeakWORDGAP,wordgap,0);
break;
case 'v':
strncpy0(voicename,param_string(&p),sizeof(voicename));
break;
case 'w':
option_waveout=1;
strncpy0(wavefile,param_string(&p),sizeof(wavefile));
break;
case '-':
strcpy(command,param_string(&p));
strncpy0(command,param_string(&p),sizeof(command));
if(memcmp(command,"compile=",8)==0)
{
CompileDictionary(NULL,&command[8],NULL,NULL);
@@ -878,7 +782,7 @@ void command_line(char *arg_string, int wait)
else
if(strcmp(command,"help")==0)
{
printf("\n%s\n%s",version_string,help_text);
printf("\nspeak text-to-speech: %s\n%s",version_string,help_text);
return;
}
else
@@ -912,7 +816,7 @@ void command_line(char *arg_string, int wait)
}
}
LoadVoice(voicename,0);
SetVoiceByName(voicename);
if((filename[0]==0) && (p[0]=='\r'))
{
@@ -1212,7 +1116,7 @@ _kernel_oserror *user_init(char *cmd_fail, int podule_base, void *pw)
}
ReadVoiceNames();
LoadVoice("default",0);
SetVoiceByName("default");
for(param=0; param<N_SPEECH_PARAM; param++)
param_stack[0].parameter[param] = param_defaults[param];
+7 -21
View File
@@ -20,8 +20,6 @@
// Header file for RISCOS build
#define N_PEAKS 9
#define N_MARKERS 7
// comment this out when compiling the "speak" process
//#define SPECT_EDITOR
@@ -43,32 +41,20 @@ typedef struct {
} MNEM_TAB;
int LookupMnem(MNEM_TAB *table, char *string);
#ifdef PLATFORM_WINDOWS
#define N_PATH_HOME 220
#else
#define N_PATH_HOME 150
#endif
typedef struct {
short pkfreq;
short pkheight;
short pkwidth;
short pkright;
} peak_t;
extern char path_home[N_PATH_HOME]; // this is the espeak-data directory
typedef struct {
short frflags;
unsigned char length;
unsigned char rms;
short ffreq[9];
unsigned char fheight[9];
unsigned char fwidth[6]; // width/4
unsigned char fright[6]; // width/4
} frame_t;
extern void strncpy0(char *to,const char *from, int size);
extern int GetFileLength(const char *filename);
extern char *Alloc(int size);
extern void Free(void *ptr);
extern void strncpy0(char *to,const char *from, int size);
extern char path_home[];
extern const char *version_string;
extern const int version_phdata;
const int wcslen(const wchar_t *str);
+1
View File
@@ -4,6 +4,7 @@ int iswalpha(int c);
int iswdigit(int c);
int iswalnum(int c);
int towlower(int c);
int towupper(int c);
int iswupper(int c);
int iswlower(int c);
int iswspace(int c);
Binary file not shown.
+1 -1
View File
@@ -3,7 +3,7 @@
initialisation-code: user_init
title-string: Speak
help-string: Speak 3.25 Text to Speech (32 bit)
help-string: Speak 3.30 Text to Speech (32 bit)
swi-chunk-base-number: &4ad80
swi-handler-code: swi_handler
+14 -6
View File
@@ -2,7 +2,9 @@
[Setup]
AppName=eSpeak
AppVerName=eSpeak version 1.29.03
AppVerName=eSpeak version 1.30
AppCopyright=Licensed under GNU General Public License version 3. (See file License.txt for details).
WindowVisible=yes
DefaultDirName={pf}\eSpeak
DefaultGroupName=eSpeak
@@ -22,7 +24,7 @@ Name: "{app}\espeak-data\mbrola"
Source: "espeak_sapi.dll"; DestDir: "{app}"; Flags: regserver promptifolder replacesameversion
Source: "TTSApp.exe"; DestDir:"{app}"
Source: "espeak-data\*"; DestDir: "{app}\espeak-data"; Flags: recursesubdirs
Source: "dictsource2\*"; DestDir: "{app}\dictsource"
Source: "dictsource\*"; DestDir: "{app}\dictsource"; Flags: recursesubdirs
Source: "docs\*"; DestDir: "{app}\docs"; Flags: recursesubdirs
Source: "command_line\*"; DestDir: "{app}\command_line"
Source: "Readme.txt"; DestDir: "{app}"; Flags: isreadme
@@ -99,13 +101,15 @@ var
begin
lang_main := language and $3ff;
Result := 'af';
Result := 'en';
// translation from microsoft codes to language codes
case lang_main of
$04: Result := 'zh';
$05: Result := 'cs';
$07: Result := 'de';
$08: Result := 'el';
$09: Result := 'en';
$0a: Result := 'es';
$0b: Result := 'fi';
$0c: Result := 'fr';
@@ -185,6 +189,7 @@ begin
'sv': value := $41d;
'sw': value := $441;
'vi': value := $42a;
'zh': value := $804;
// mbrola voices
'-af': value := $436;
'-br': value := $416;
@@ -356,9 +361,12 @@ begin
UIVoice := VoiceFromLanguage(UILanguage);
// Set initial values (optional)
Page.Values[0] := 'en';
Page.Values[1] := UIVoice;
Page.Values[2] := 'en-r';
Page.Values[0] := UIVoice;
Page.Values[1] := UIVoice+'+f2';
if UIVoice = 'en' then
Page.Values[2] := 'en-r'
else
Page.values[2] := 'en';
end;
+2 -2
View File
@@ -1,7 +1,7 @@
[Setup]
AppName=eSpeakEdit
AppVerName=eSpeakEdit version 1.29
AppVerName=eSpeakEdit version 1.30
DefaultDirName={pf}\eSpeak
DefaultGroupName=eSpeak
OutputBaseFilename=setup_espeakedit
@@ -12,7 +12,7 @@ ShowLanguageDialog=auto
[Files]
Source: "espeakedit.exe"; DestDir: "{app}"
Source: "dictsource\*"; DestDir: "{app}\dictsource"; Flags: recursesubdirs
;Source: "dictsource\*"; DestDir: "{app}\dictsource"; Flags: recursesubdirs
Source: "espeakedit\*"; DestDir: "{app}\espeakedit"; Flags: recursesubdirs
Source: "phsource\*"; DestDir: "{app}\phsource"; Flags: recursesubdirs
Source: "docs\*"; DestDir: "{app}\docs"; Flags: recursesubdirs
+5 -5
View File
@@ -5,8 +5,8 @@
//
VS_VERSION_INFO VERSIONINFO
FILEVERSION 1,0,24,00
PRODUCTVERSION 1,0,24,00
FILEVERSION 1,0,30,00
PRODUCTVERSION 1,0,30,00
FILEFLAGSMASK 0x3fL
#ifdef _DEBUG
FILEFLAGS 0x1L
@@ -24,14 +24,14 @@ BEGIN
VALUE "Comments", "\0"
VALUE "CompanyName", "Jonathan Duddington\0"
VALUE "FileDescription", "Text to Speech\0"
VALUE "FileVersion", "1, 0,24,00\0"
VALUE "InternalName", "espeak1.24\0"
VALUE "FileVersion", "1, 0,30,00\0"
VALUE "InternalName", "espeak1.30\0"
VALUE "LegalCopyright", "Copyright (c) Jonathan Duddington\0"
VALUE "LegalTrademarks", "\0"
VALUE "OriginalFilename", "espeak_sapi.dll\0"
VALUE "PrivateBuild", "\0"
VALUE "ProductName", "espeak TTS\0"
VALUE "ProductVersion", "1, 0, 24, 00\0"
VALUE "ProductVersion", "1, 0, 30, 00\0"
VALUE "SpecialBuild", "\0"
END
END