[1.26.05] Fix count of upper-case letters in a word to consider multibyte characters.

Mbrola translations, option to not match across a word boundary.
Mbrola translations, allow check of lengthen indicator [:].
espeakedit, add operation to apply a voice's formants modifications to a keyframe.


git-svn-id: https://espeak.svn.sourceforge.net/svnroot/espeak/trunk@47 d46cf337-b52f-0410-862d-fd96e6ae7743
This commit is contained in:
jonsd
2007-06-14 18:41:52 +00:00
parent 2dbbff737b
commit 188b0d7c11
20 changed files with 220 additions and 131 deletions
+5
View File
@@ -271,6 +271,7 @@ Translator *SelectTranslator(const char *name)
case L('h','r'): // Croatian
case L('s','r'): // Serbian
{
static const wchar_t replace_chars_hr[] =
{0x430,0x431,0x446,0x45b,0x447,0x434,0x452,0x45f,
@@ -389,6 +390,10 @@ Translator *SelectTranslator(const char *name)
tr = new Translator();
tr->charset_a0 = charsets[4]; // ISO-8859-4, includes a,e,i,o,u-macron
tr->langopts.stress_rule = 2;
tr->langopts.stress_flags = 0x21;
tr->langopts.unstressed_wd1 = 0;
tr->langopts.unstressed_wd2 = 2;
tr->langopts.param[LOPT_DIERESES] = 1;
}
break;