Signed-off-by: Attila Hammer <[email protected]>

Hi Boys,

Since between 2010 and 2012 year, when I created original hu_list and
hu_rules files, me not have more time to extending this rule set with
more new pronounce rules.
I consulted previous with @alex19EP and promise this larger pull request
after february 20, but I simple forgot opening this PR after Liblouis
3.29.0 release publication in march 4TH. :-):-)
SO:
* In hu_list file not happened larger changes, I doed only fiwe
optimizations. This change I stored a small first commit.
* The hu_rules file contains new or partially modified pronounce rules
for example with cs, ccs, ggy, nny, sz, ssz letters, and contains end of
hu_rules file a critical need modification with plus and minus simbols
handling related, with affects the Android and perhaps iOS versions.
Previous, with oldest Espeak-ng releases hungarian users reported me
with if using Espeak-ng, not hear temperature value numbers the plus and
minus simbols (for example the degree have -1 and +1 celsius), the users
hear only 1 and 5 numbers.
I am not detected this, because I am using a very old Symbian cell
phone, and in Linux using Orca with most punctuation level with
Speech-dispatcher and Espeak-ng.

@alex19EP, the second commit is very large, estimated the diff output is
1236 lines. :-):-)
The second commit contains only hungarian language related word
pronounce modifications with already tested and listened literal texts,
so not will be producing the second commit with regressions the other
languages related rules.

If this is possible, please merge this two commits the proper branches
into the Espeak-ng source tree branches, and please publicate this large
rule extending related modification the new stable Espeak-ng release
with wrote me in your e-mail.
Forgive my inattention for not publishing this larger pull request to
you sooner.

I have only one question:
After Espeak-ng updated the ruleset related files (the hu_list,
hu_rules, and binary compiled hungarian dictionary files the stable
version), in Piper project need replace hungarian dictionary files, or
this is happens automatically with a cooperation?
If I known right, Piper TTS uses perhaps original or a modified
Espeak-ng with phoneme processing related.
Looks following README content:
https://github.com/rhasspy/piper

Thank you the cooperation, and sorry the late pull request dropping,

Attila

This larger commit implement new previous missing more word pronounce rules,
and fixing the minus and plus punctuation simbols before and after numbers,
for example Android and iPhone version.
This rule changes provide more better hungarian pronounce quality, based with
literary books.

Signed-off-by: Attila Hammer <[email protected]>

Signed-off-by: Attila Hammer <[email protected]>

This PR adds more Washington State municipalities. Pronunciations
verified from personal experience and additionally by Wikipedia's IPA.

The number 1000000000 tries to have a voiceless velar
fricative, I think, which is NOT how sextillion is pronounced in
English.
```
espeak-ng -q -x 1000000000
w'0n sExt'Ili@n
```

$(srcdir) was missing on includes for src/include/espeak-ng/espeak-ng.h

$(srcdir) was missing on includes for src/include/espeak-ng/espeak-ng.h

Next PR removing rules which ,make sense for English, but should not be
present in the Polish language files.

So we can run tests parallel with cmake, reducing test time and IO
overhead

Fixes: #1823, #1824, #1825, #1826, #1827

- Add crash test and vectors provided by @SEU-SSL
- Disallow dummy/null voice load (that causes incorrect translator
initialization)
- Fix empty `phondata` file load (that causes unitialized memory access)
- Limit max word length for RemoveEnding (causes buffer overflow)
- Limit punctlist initialization from embedded commands (buffer
overflow)
- Fix unitialized pitch in wavegen (DBZ and indexing problems)
- Properly zeroize stack variables before use in TranslateClause and
SetWordStress

TODO (in nextup PR): add & fix more vectors from fuzzer.