mjohnson/espeak-ng
1
0
Fork 0
mirror of https://github.com/rhasspy/espeak-ng.git synced 2026-11-01 21:58:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'master' into android

Browse Source
This commit is contained in:
Reece H. Dunn
2013-04-29 10:29:51 +01:00
parent f57de435ac 0122c41c4d
commit 4236dd282e
95 changed files with 4592 additions and 2547 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/numbers.cpp
+148 -9
View File
@@ -582,7 +582,103 @@ void LookupLetter(Translator *tr, unsigned int letter, int next_byte, char *ph_b
} // end of LookupLetter
static const char *hex_letters[] = {"'e:j","b'i:","s'i:","d'i:","'i:","'Ef"}; // using phonemes available to all languages
// unicode ranges for non-ascii digits 0-9
static const int number_ranges[] = {
0x660, 0x6f0, // arabic
0x966, 0x9e6, 0xa66, 0xae6, 0xb66, 0xbe6, 0xc66, 0xce6, 0xd66, // indic
0xe50, 0xed0, 0xf20, 0x1040, 0x1090,
0 }; // these must be in ascending order
int NonAsciiNumber(int letter)
{//============================
// Change non-ascii digit into ascii digit '0' to '9', (or -1 if not)
const int *p;
int base;
for(p=number_ranges; (base = *p) != 0; p++)
{
if(letter < base)
break; // not found
if(letter < (base+10))
return(letter-base+'0');
}
return(-1);
}
#define L_SUB 0x4000 // subscript
#define L_SUP 0x8000 // superscript
static const char *modifiers[] = {NULL, "_sub", "_sup", NULL};
// this list must be in ascending order
static unsigned short derived_letters[] = {
0x00aa, 'a'+L_SUP,
0x00b2, '2'+L_SUP,
0x00b3, '3'+L_SUP,
0x00b9, '1'+L_SUP,
0x00ba, 'o'+L_SUP,
0x02b0, 'h'+L_SUP,
0x02b1, 0x266+L_SUP,
0x02b2, 'j'+L_SUP,
0x02b3, 'r'+L_SUP,
0x02b4, 0x279+L_SUP,
0x02b5, 0x27b+L_SUP,
0x02b6, 0x281+L_SUP,
0x02b7, 'w'+L_SUP,
0x02b8, 'y'+L_SUP,
0x02c0, 0x294+L_SUP,
0x02c1, 0x295+L_SUP,
0x02e0, 0x263+L_SUP,
0x02e1, 'l'+L_SUP,
0x02e2, 's'+L_SUP,
0x02e3, 'x'+L_SUP,
0x2070, '0'+L_SUP,
0x2071, 'i'+L_SUP,
0x2074, '4'+L_SUP,
0x2075, '5'+L_SUP,
0x2076, '6'+L_SUP,
0x2077, '7'+L_SUP,
0x2078, '8'+L_SUP,
0x2079, '9'+L_SUP,
0x207a, '+'+L_SUP,
0x207b, '-'+L_SUP,
0x207c, '='+L_SUP,
0x207d, '('+L_SUP,
0x207e, ')'+L_SUP,
0x207f, 'n'+L_SUP,
0x2080, '0'+L_SUB,
0x2081, '1'+L_SUB,
0x2082, '2'+L_SUB,
0x2083, '3'+L_SUB,
0x2084, '4'+L_SUB,
0x2085, '5'+L_SUB,
0x2086, '6'+L_SUB,
0x2087, '7'+L_SUB,
0x2088, '8'+L_SUB,
0x2089, '9'+L_SUB,
0x208a, '+'+L_SUB,
0x208b, '-'+L_SUB,
0x208c, '='+L_SUB,
0x208d, '('+L_SUB,
0x208e, ')'+L_SUB,
0x2090, 'a'+L_SUB,
0x2091, 'e'+L_SUB,
0x2092, 'o'+L_SUB,
0x2093, 'x'+L_SUB,
0x2094, 0x259+L_SUB,
0x2095, 'h'+L_SUB,
0x2096, 'k'+L_SUB,
0x2097, 'l'+L_SUB,
0x2098, 'm'+L_SUB,
0x2099, 'n'+L_SUB,
0x209a, 'p'+L_SUB,
0x209b, 's'+L_SUB,
0x209c, 't'+L_SUB,
0,0};
static const char *hex_letters[] = {"'e:j","b'i:","s'i:","d'i:","'i:","'ef"}; // names, using phonemes available to all languages
int TranslateLetter(Translator *tr, char *word, char *phonemes, int control)
{//=========================================================================
@@ -596,15 +692,18 @@ int TranslateLetter(Translator *tr, char *word, char *phonemes, int control)
int letter;
int len;
int ix;
int c;
char *p2;
char *pbuf;
const char *modifier;
ALPHABET *alphabet;
int al_offset;
int al_flags;
int language;
int number;
int phontab_1;
int speak_letter_number;
char capital[20];
char capital[30];
char ph_buf[80];
char ph_buf2[80];
char ph_alphabet[80];
@@ -626,21 +725,61 @@ int TranslateLetter(Translator *tr, char *word, char *phonemes, int control)
if(control & 2)
{
// include CAPITAL information
if(iswupper(letter))
if(iswupper2(letter))
{
Lookup(tr, "_cap", capital);
}
}
letter = towlower2(letter);
LookupLetter(tr, letter, word[n_bytes], ph_buf, control & 1);
if(ph_buf[0] == 0)
{
// is this a subscript or superscript letter ?
for(ix=0; (c = derived_letters[ix]) != 0; ix+=2)
{
if(c > letter)
break;
if(c == letter)
{
c = derived_letters[ix+1];
letter = c & 0x3fff;
if((modifier = modifiers[c >> 14]) != NULL)
{
Lookup(tr, modifier, capital);
if(capital[0] == 0)
{
capital[2] = SetTranslator2("en"); // overwrites previous contents of translator2
Lookup(translator2, modifier, &capital[3]);
if(capital[3] != 0)
{
capital[0] = phonPAUSE;
capital[1] = phonSWITCH;
len = strlen(&capital[3]);
capital[len+3] = phonSWITCH;
capital[len+4] = phontab_1;
capital[len+5] = 0;
}
}
}
}
}
LookupLetter(tr, letter, word[n_bytes], ph_buf, control & 1);
}
if(ph_buf[0] == phonSWITCH)
{
strcpy(phonemes,ph_buf);
return(0);
}
if((ph_buf[0] == 0) && ((number = NonAsciiNumber(letter)) > 0))
{
// convert a non-ascii number to 0-9
LookupLetter(tr, number, 0, ph_buf, control & 1);
}
al_offset = 0;
al_flags = 0;
if((alphabet = AlphabetFromChar(letter)) != NULL)
@@ -766,7 +905,7 @@ int TranslateLetter(Translator *tr, char *word, char *phonemes, int control)
speak_letter_number = 1;
if(!(al_flags & AL_NO_SYMBOL))
{
if(iswalpha(letter))
if(iswalpha2(letter))
Lookup(translator, "_?A", ph_buf);
if((ph_buf[0]==0) && !iswspace(letter))
@@ -784,8 +923,8 @@ int TranslateLetter(Translator *tr, char *word, char *phonemes, int control)
speak_letter_number = 0;
}
if((ph_alphabet[0] != 0) && speak_letter_number)
ph_buf[0] = 0; // don't speak "letter" if we speak alphabet name
// if((ph_alphabet[0] != 0) && speak_letter_number)
// ph_buf[0] = 0; // don't speak "letter" if we speak alphabet name
if(speak_letter_number)
{
@@ -2216,8 +2355,8 @@ static int TranslateNumber_1(Translator *tr, char *word, char *ph_out, unsigned
if((tr->langopts.numbers & NUM_NOPAUSE) && (next_char == ' '))
utf8_in(&next_char,p);
if(!iswalpha(next_char) && (thousands_exact==0))
// if(!iswalpha(next_char) && !((wtab[thousandplex].flags & FLAG_HYPHEN_AFTER) && (thousands_exact != 0)))
if(!iswalpha2(next_char) && (thousands_exact==0))
// if(!iswalpha2(next_char) && !((wtab[thousandplex].flags & FLAG_HYPHEN_AFTER) && (thousands_exact != 0)))
strcat(ph_out,str_pause); // don't add pause for 100s, 6th, etc.
}