mjohnson/espeak-ng
1
0
Fork 0
mirror of https://github.com/rhasspy/espeak-ng.git synced 2026-10-31 13:38:47 +00:00
Code Issues Packages Projects Releases Wiki Activity
4,192 Commits 1 Branch 4 Tags
08c36e61eca12bfe3887dc6fd0156d63ffdaf168
Commit Graph
1678 Commits
Author SHA1 Message Date
Valdis Vitolins e9c66a4558 Support up to 300 voices/languages to make some space 2019-09-05 22:08:33 +03:00
Valdis Vitolins 1f5b774cce Add warning when number of voices reaches limit 2019-09-05 22:04:45 +03:00
Reece H. Dunn 07f9a9ce3f Remove the unused variable in voices.c. 2019-09-04 19:38:34 +01:00
Reece H. Dunn f58f665336 Support a maximum of 275 voices/languages. 
The new voices added take the voice list over 250 voices/languages.
 2019-09-04 19:35:40 +01:00
Reece H. Dunn 68939f24bf Merge remote-tracking branch 'abbr/zhy-dict' 2019-07-30 19:05:16 +01:00
unknown da09648b45 Fix path for hy-arevmda (replace hy-arevmda with hyw, now windows installer works again). 2019-07-18 00:04:15 -04:00
Valdis Vitolins 15f8741bce Merge pull request #607 2019-04-21 15:48:38 +03:00
Anh Hoang Nguyen 2504455ed9 vi: modify phoneme data of e 2019-03-20 14:41:02 +07:00
Anh Hoang Nguyen 38abb79ecd vi: modify data of phoneme u and i 2019-03-20 14:18:33 +07:00
golubovsky 009b6b5d14 Finished the voice load code. Dumping voice def does not seem possible 
because it is not stored in a single place.

Signed-off-by: golubovsky <golubovsky@gmail.com>
 2019-02-23 22:51:32 -05:00
golubovsky 439f7f32d5 First part of the patch: add option --load 
Option --load makes espeak-ng consider the voice name provided a name of
a file relative to the current directory.

Signed-off-by: golubovsky <golubovsky@gmail.com>
 2019-02-21 22:50:09 -05:00
Samuel Thibault 04a8a619b6 Fix termination test in event processing 
e13b602 ("Avoid using pthread_cancel") introduced a bogus termination test,
this fixes it.

Fixes #572
 2018-12-19 15:54:21 +01:00
abbr 3e2467e65d compile common zhy character dict by default 2018-12-04 17:29:47 -08:00
Samuel Thibault e13b602ce0 Avoid using pthread_cancel 
If a thread is cancelled while it holds a mutex, other threads waiting for
the mutex will never be woken, even if destroying the mutex, thus
leading to a deadlock.

Better just make the thread loops have a proper exit case than risking
the tricky semantic of pthread_cancel.
 2018-11-13 01:55:17 +01:00
Reece H. Dunn 919f3240cb Revert "Move handling of SetLetterVowel() to language files" 
This reverts commit 6042eb8d1e.

The change causes an issue with `pt-BR` and possibly other languages
and voices.
 2018-10-08 07:38:56 +01:00
Reece H. Dunn 348e7ecbd6 Merge remote-tracking branch 'jaacoppi/manual' 2018-10-06 10:02:38 +01:00
Juho Hiltunen 520a30e0b0 code cleanup: use existing defines espeakRATE_MINIMUM, espeakRATE_NORMAL and espeakRATE_MAXIMUM. 2018-10-06 11:34:25 +03:00
Juho Hiltunen 3da5af52aa man: fix incorrect default speed value 
Fixes #555.
 2018-10-06 11:00:08 +03:00
Valdis Vitolins 6bdb2ee6cd ar: in MBROLA qaf is not considered 'thick' consonant 2018-09-30 17:45:25 +03:00
Valdis Vitolins 26b067a68b ar: move default stress rule from code to config file 2018-09-11 18:02:27 +03:00
Reece H. Dunn 9d901e48be shn: support numbers for 100, 1000, and 10000 2018-08-25 22:02:08 +01:00
Reece H. Dunn 61672f5e24 Use defines for the different number breaking systems to improve readability. 2018-08-25 21:51:26 +01:00
Reece H. Dunn 273b8e2a1f en: support numbers upto a hundred nonillion 2018-08-25 19:53:01 +01:00
Reece H. Dunn 9638750579 Always flush stdout when reading stdin line by line. 2018-08-22 07:42:27 +01:00
Reece H. Dunn 1c60fb7f62 Don't use STRESSPOSN_1L for thousands_sep in the Slovak/Czech language setup. 2018-08-19 19:31:19 +01:00
Reece H. Dunn c6ac526847 When printing phonemes, don't add a space at the start of a sentence or clause. 2018-08-19 13:11:18 +01:00
Reece H. Dunn 65186c07df Preserve the sourceix property of a deleted phonSWITCH phoneme. 2018-08-19 10:11:05 +01:00
Reece H. Dunn cf6d14783c Preserve the sourceix property of a deleted phoneme for replaced phonemes. 2018-08-19 09:35:03 +01:00
Reece H. Dunn 8e13f7147c Add constants for use with PHONEME_LIST.newword. 2018-08-19 08:29:09 +01:00
Reece H. Dunn 86bbc257b0 Support matching any length strings in the replacement rules. 2018-07-24 22:32:39 +01:00
Reece H. Dunn 98e9122dfc FindReplacementChars: Pass in the source buffer (next characters) instead of next_in. 2018-07-24 22:04:12 +01:00
Reece H. Dunn 4fbcda9c2a FindReplacementChars: Use an nc (next character) variable. 2018-07-24 22:00:25 +01:00
Reece H. Dunn cacc212d4b FindReplacementChars: Rename uc to fc. 2018-07-24 21:57:54 +01:00
Reece H. Dunn a9d4bdd7f7 Make ignore_next into ignore_next_n to support ignoring multiple next characters. 2018-07-24 21:30:47 +01:00
Reece H. Dunn 252f5772ae Simplify printing the replace message. 2018-07-23 22:28:09 +01:00
Reece H. Dunn 55c64036e0 Use UTF-8 strings in replace rules, instead of a packed UTF-16 pair. 2018-07-23 22:12:20 +01:00
Reece H. Dunn 0e91fcbc04 Don't use pw when reading the replacement data. 2018-07-23 21:15:05 +01:00
Reece H. Dunn 424f705525 Revert the new (broken) replacement rule logic. 
The replacement tests for bs, hr, and sr are no longer marked as
broken as they work using the old code. The mk tests keep the
broken annotation, as they don't work in the old code either.

This reverts commit 801a8d197c.
This reverts commit 64d5701e5e.
This reverts commit 3b51ebf617.
This reverts commit 1fd235d2c0.
This reverts commit 9f0667de86.
 2018-07-23 19:38:46 +01:00
Reece H. Dunn bba7069cb3 issue #520: Use .replace rules in the language rule files for Cyrillic to Latin 2018-07-21 14:09:04 +01:00
Reece H. Dunn 32ab396ea1 Don't declare the Arabic letter strings using array syntax. 2018-07-19 20:21:09 +01:00
Reece H. Dunn bfe184c641 Make PrepareLetters actually set the letter bits, and rename to SetLetterBitsUTF8. 2018-07-19 20:19:15 +01:00
Reece H. Dunn 7f42bd39b4 Move PrepareLetters next to SetLetterBitsRange. 2018-07-19 20:03:23 +01:00
Reece H. Dunn 22ee347234 Use script name prefixes in the Set[Script]Letters functions for group bit lists. 2018-07-19 20:00:47 +01:00
Reece H. Dunn 36c6727e90 issue #518: Add the 'е є ї' letters to the Y group in Translate_Russian, not SetCyrillicLetters. 2018-07-19 19:49:28 +01:00
Reece H. Dunn 17dac5ea53 Add a comment to SetCyrillicLetters describing which languages it applies to. 2018-07-19 19:19:03 +01:00
Reece H. Dunn e7e59f99d3 Fix the warnings in PrepareLetters. 2018-07-16 07:51:20 +01:00
Valdis Vitolins 6b57e97283 ar: set default stress on antepenultimate syllable 2018-07-15 18:41:50 +03:00
Valdis Vitolins c92e70183c ar: allow diacritics before normal letters as Arabic characters 2018-07-15 18:33:20 +03:00
Valdis Vitolins 1cf7549392 Prepare Arabic letters using PrepareLetters() function 
PrepareLetters() function also could be used to set up letter groups A, B, C, etc. from string
and further could be loaded from configuration file
 2018-07-15 18:26:28 +03:00
Reece H. Dunn 801a8d197c TranslateChars: Supress an unsigned to signed cast 2018-07-14 08:10:41 +01:00