Date   

Re: possible bug in the linux version of espeak ng

Reece H. Dunn
 

On Fri, Jan 20, 2017 at 01:18 am, kendell clark wrote:

hi all

I'm a co developer of the sonar linux distribution, which you can get at
http://www.sonargnulinux.com. We're in the process of switching the
distribution we're based on, from manjaro to antergos. I've re built the
espeak ng package for antergos, since arch hasn't yet switched from
espeak to espeak ng. It built just fine, but when I launch espeak hi to
make sure everything works, I get the following error. Cannot process
file /home/kendell/phontab.

Hi Kendell,


Thanks for the error report. I have pushed a fix for this now.

Kind regards,

Reece


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

1 New Commit:

[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
d42ba9395926: Don't use the HOME directory as the espeak-ng-data path.

Modified: src/libespeak-ng/speech.c


2 New Commits:

[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
b3daacc79585: Document the Kirshenbaum IPA features.

Added: docs/phoneme_features.md


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
68234bdbb303: Merge remote-tracking branch 'rhdunn/master'

Added: docs/phoneme_features.md


[espeak-ng:master] reported: Manipuri numbers are pronounced wrongly #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Comment on Issue #100 Manipuri numbers are pronounced wrongly
By ValdisVitolins:

Even though eSpeakNG has sound inventory (phoneme list) for Manipuri language, I couldn't find pronunciation rules for it. Closest language I could find is Bengali (bn_list and bn_rules files). Is this appropriate language to look at?


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng/espeak-ng] Pull request opened by ValdisVitolins:

#200 Fix for issue #69 Speaking '1,,2' in German only says 'eins'.


[espeak-ng/espeak-ng] Pull request updated by ValdisVitolins:

#200 Fix for issue #69 Speaking '1,,2' in German only says 'eins'.
I tried to make espeak-ng say with comma, but unsuccesfully. This implementation says ` _'aIns tsv'aI` without comma, similarly to consecutive dots in number.


[espeak-ng:master] New Comment on Issue #155 Letter group of rules should allow to add "empty string" as one of possible options
By ValdisVitolins:

As remaining issue with rule translation is documented in issue #196, this one can be considered as solved.


[espeak-ng:master] Issue #155 Letter group of rules should allow to add "empty string" as one of possible options closed by ValdisVitolins.


possible bug in the linux version of espeak ng

kendell clark
 

hi all

I'm a co developer of the sonar linux distribution, which you can get at http://www.sonargnulinux.com. We're in the process of switching the distribution we're based on, from manjaro to antergos. I've re built the espeak ng package for antergos, since arch hasn't yet switched from espeak to espeak ng. It built just fine, but when I launch espeak hi to make sure everything works, I get the following error. Cannot process file /home/kendell/phontab. I've tested this on my wife mellisa's machine, and the error is instead cannot process /home/mellisa/phontab I'm not sure what's going on, but it sounds like maybe espeak ng is looking for it's phontab file and not finding it. Would someone mind looking into this? I can file an issue on git hub if you want, but I thought I'd post it hear in case this is a packaging issue. I maintain the package for sonar, and it's always possible I made a mistake somewhere, although I haven't changed the build script in a while. Has anyone else experienced this?

Thanks

Kendell Clark


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by sukiletxe:
#130 Spanishç

The ç in Spanish is pronounced as /k/, and it should be pronounced as /s/.


[espeak-ng:master] New Issue Created by Christianlm:
#129 Spanish: vowels and diphthongs

I'm interested to open this issue and to solve it.

In ph_spanish is used the same value for each vowel. I think that is incorrect and it's needed different length in the various vowels and for unstressed vowels.

The phoneme o and phoneme O statements are the same, using the same phoneme file.

Same way for phoneme e and phoneme E, it would be better to create instructions for unstressed /o/ and unstressed /e/.

Also, eSpeak-ng pronounces an incorrect diphthong /oi/. I'm working to create two diphthongs for spanish /oi/ with the /o/ is more central and open.

Thoughts?

I will push experimental phonemes and statements for spanish language here:

https//github.com/Christianlm/SpanishOnEspeak-ng.git

Thanks.


[espeak-ng:master] New Issue Created by ValdisVitolins:
#141 Documentation update: espeak vs espeak-ng

Probably this should be included from Reece's e-mail

The espeak-ng project works as a drop-in replacement for espeak, with a few caveats. More specifically: 1. The installation creates compatibility symlinks mapping espeak, speak and libespeak.la to their espeak-ng equivalents. These are optional if you just want to install espeak-ng and have an espeak package with the espeak code and an espeak-ng-compatibility package that installs the symlinks. 2. The command-line interface to espeak-ng and speak-ng are compatible, and will remain so, with espeak. 3. The C API to libespeak-ng.so will remain API and ABI compatible with libespeak.so. The only change I have made and intend to make to speak_lib.h is a change to the ESPEAK_API macro when building on Windows. All new APIs are being added to espeak-ng/espeak_ng.h. 4. The espeak-data files have been modified slightly. Currently espeak-ng can read espeak data, but espeak cannot read espeak-ng data. I can make it so that espeak-ng uses a different voice data so that both espeak and espeak-ng can be installed on the same machine without problems. 5. The espeak-ng project does not include espeakedit. The voice data is built using the espeak-ng command line itself. Currently, to create and edit the voice data, you need espekaedit from the espeak project.

Some other things to be aware of: 1. The project uses a standard autotools build process. 2. The project is dependent on the system's installation of the sonic library. 3. The project uses an external pcaudiolib project (https://github.com/espeak-ng/pcaudiolib) for audio support.


[espeak-ng:master] New Issue Created by rhdunn:
#69 Speaking '1,,2' in German only says 'eins'.

The following is spoken given the input 1,,2:

$src/espeak-ng -vde -x "1,,2"
_'aIns
$src/espeak-ng -ven -x "1,,2"
w'0n t'u:

The German output should be:

_'aIns tsv'aI

Source: https://github.com/nvaccess/nvda/issues/4956


[espeak-ng:master] New Issue Created by rhdunn:
#69 Speaking '1,,2' in German only says 'eins'.

The following is spoken given the input 1,,2:

$src/espeak-ng -vde -x "1,,2"
_'aIns
$src/espeak-ng -ven -x "1,,2"
w'0n t'u:

The German output should be:

_'aIns tsv'aI

Source: https://github.com/nvaccess/nvda/issues/4956


[espeak-ng:master] New Issue Created by rhdunn:
#24 Support selecting alternative voice data

The voice data is the phoneme tables and intonations. This would allow selecting a different location for this data: 1. if the location is a directory, it loads the files separately from that directory (like is done currently, but allowing a different location than the espeak-data path); 2. if the location is a file, this loads a combined data file that contains voice metadata and the phonemetable and intonation data files.

Access to this functionality should be provided by the C API and command-line interface.


[espeak-ng:master] New Issue Created by rhdunn:
#24 Support selecting alternative voice data

The voice data is the phoneme tables and intonations. This would allow selecting a different location for this data: 1. if the location is a directory, it loads the files separately from that directory (like is done currently, but allowing a different location than the espeak-data path); 2. if the location is a file, this loads a combined data file that contains voice metadata and the phonemetable and intonation data files.

Access to this functionality should be provided by the C API and command-line interface.


[espeak-ng:master] Label added to issue #24 Support selecting alternative voice data by rhdunn.


[espeak-ng:master] Issue #24 Support selecting alternative voice data closed by rhdunn.


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

2 New Commits:

[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
0367abf98f2d: Support loading phoneme data files relative to the phsrc path.

Modified: src/libespeak-ng/compiledata.c


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
773da20c16dd: Support using home (data) paths other than 'espeak-ng-data'.

Modified: CHANGELOG.md
Modified: src/libespeak-ng/speech.c


[espeak-ng:master] Label added to issue #199 Language analysis improvements by ValdisVitolins.


[espeak-ng:master] Label added to issue #199 Language analysis improvements by ValdisVitolins.


[espeak-ng:master] New Comment on Issue #69 Speaking '1,,2' in German only says 'eins'.
By ValdisVitolins:

I'm working on this and (hopefully) will provide patch tomorrow.


[espeak-ng:master] New Issue Created by ValdisVitolins:
#195 Different non-Latin diacritized consonants are wrongly considered as one doubled

if following rule a (C% u, then it passes also for 'aņša'. As workaround one can use character group (e.g. .L01 ņ and .L02 š) and ... (L01% ... (L02% rules instead.


[espeak-ng:master] New Issue Created by ValdisVitolins:
#195 Different non-Latin diacritized consonants are wrongly considered as one doubled

if following rule a (C% u, then it passes also for 'aņša'. As workaround one can use character group (e.g. .L01 ņ and .L02 š) and ... (L01% ... (L02% rules instead.


[espeak-ng:master] New Issue Created by sthibaul:
#148 bad pronunciation of "unary"

Hello, From http://bugs.debian.org/830731 :

“ espeak pronunces "unary" as: [[Vn'A@r-i]]

I believe the correct pronunciation is: [[j'un@ri]] ”

This happens with espeak-ng too.


[espeak-ng:master] New Issue Created by sthibaul:
#149 EWMH, UCS, UTF, UUID acronyms

Hello,

From http://bugs.debian.org/830730 :

“ These acronyms are currently not pronounced correctly: EWMH, UCS, UTF, UUID ”

A proposed patch was submitted: espeak-abbrev.diff.txt


[espeak-ng:master] New Issue Created by burakyuksek:
#152 Turkish pronunciation problems

Hello, I have a few problems to report with the Turkish language: 1. Words that repeat a silent letter twice has the first instance of the letter muted. For example, in the word gitti, the first t needs to be muted. 2. The letters x, q and w are not pronounced correctly. X needs to be pronounced as iks rather than eks, w should be pronounced as çift v rather than duble v because duble is not a Turkish word, and the letter q needs to be pronounced as kyu rather than kve. 3. Words like sertifika and dakika should have the k a bit thicker.


[espeak-ng:master] New Issue Created by burakyuksek:
#152 Turkish pronunciation problems

Hello, I have a few problems to report with the Turkish language: 1. Words that repeat a silent letter twice has the first instance of the letter muted. For example, in the word gitti, the first t needs to be muted. 2. The letters x, q and w are not pronounced correctly. X needs to be pronounced as iks rather than eks, w should be pronounced as çift v rather than duble v because duble is not a Turkish word, and the letter q needs to be pronounced as kyu rather than kve. 3. Words like sertifika and dakika should have the k a bit thicker.


[espeak-ng:master] New Issue Created by michaelDCurran:
#167 endian.h missing on Windows

Recently, spect.c has started including endian.h. This does not exist on Windows, thus we can no longer compile eSpeak for the NVDA project. Broke in commit 832058167513670fa9fe23e640ece643751b125b


[espeak-ng:master] New Issue Created by michaelDCurran:
#167 endian.h missing on Windows

Recently, spect.c has started including endian.h. This does not exist on Windows, thus we can no longer compile eSpeak for the NVDA project. Broke in commit 832058167513670fa9fe23e640ece643751b125b


[espeak-ng:master] New Issue Created by sthibaul:
#171 monotonous speech by default when used as a lib?

Hello,

As reported on http://bugs.debian.org/848008 , switching brltty to espeak-ng (without source code change) makes the speech very monotonous, which makes it unusable for some work usage. I'm afraid that if it's not fixed soon I'll have to revert to espeak in Debian.

Samuel


[espeak-ng:master] New Issue Created by sthibaul:
#171 monotonous speech by default when used as a lib?

Hello,

As reported on http://bugs.debian.org/848008 , switching brltty to espeak-ng (without source code change) makes the speech very monotonous, which makes it unusable for some work usage. I'm afraid that if it's not fixed soon I'll have to revert to espeak in Debian.

Samuel


[espeak-ng:master] New Issue Created by rhdunn:
#176 Add support for Arabic

There is Arabic support in https://github.com/linuxscout/espeak-ng. This needs cleaning up to:

  1. Simplify the history to not include small changes with non-descriptive commit messages.
  2. Not add the Arabic MBROLA voice binary files to the espeak-ng repository.


[espeak-ng:master] New Issue Created by rhdunn:
#176 Add support for Arabic

There is Arabic support in https://github.com/linuxscout/espeak-ng. This needs cleaning up to:

  1. Simplify the history to not include small changes with non-descriptive commit messages.
  2. Not add the Arabic MBROLA voice binary files to the espeak-ng repository.


[espeak-ng:master] New Issue Created by midaszhou:
#180 Undefined reference to rpl_malloc

Just cross compile espeak-ng for Openwrt, Errors arise at final make stage :

image


[espeak-ng:master] New Issue Created by eeejay:
#183 Emscripten build is broken with new endian.h shim

The endian.h shim does not include a case for an emscripten build. It should just pass through to the environment's endian.h.


[espeak-ng:master] New Issue Created by eeejay:
#183 Emscripten build is broken with new endian.h shim

The endian.h shim does not include a case for an emscripten build. It should just pass through to the environment's endian.h.


[espeak-ng:master] New Issue Created by captnc:
#185 Config.h error during build (android)

Hi, i try to build espeak-ng for android : export ANDROID_HOME=... ./autogen.sh ./configure --with-gradle=... make apk-release

but i have this error : cd android && ndk-build make[1] : on entre dans le répertoire « /home/mr_pou/Travail/Espeak/espeak-master/android » [mips] Compile : ttsespeak <= compiledata.c jni/../../src/libespeak-ng/compiledata.c:2010: fatal error: 'config.h' file not found #include "config.h" ^ 1 error generated. ...

Thx


[espeak-ng:master] New Issue Created by captnc:
#185 Config.h error during build (android)

Hi, i try to build espeak-ng for android : export ANDROID_HOME=... ./autogen.sh ./configure --with-gradle=... make apk-release

but i have this error : cd android && ndk-build make[1] : on entre dans le répertoire « /home/mr_pou/Travail/Espeak/espeak-master/android » [mips] Compile : ttsespeak <= compiledata.c jni/../../src/libespeak-ng/compiledata.c:2010: fatal error: 'config.h' file not found #include "config.h" ^ 1 error generated. ...

Thx


[espeak-ng:master] New Issue Created by sthibaul:
#148 bad pronunciation of "unary"

Hello, From http://bugs.debian.org/830731 :

“ espeak pronunces "unary" as: [[Vn'A@r-i]]

I believe the correct pronunciation is: [[j'un@ri]] ”

This happens with espeak-ng too.


Github push to espeak-ng:espeak-ng #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

7 New Commits:

[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
54b1bf939c20: Ignore the 'make dist' output.

Modified: .gitignore


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
59ca718d4095: Always use 'en' as the default voice/language.

Removed: espeak-ng-data/voices/default
Modified: emscripten/js/demo.js
Modified: src/espeak-ng.c
Modified: src/libespeak-ng/voices.c


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
ca77ce928be2: CompilePhonemeData2: Inline the single use of the source parameter.

Modified: src/libespeak-ng/compiledata.c


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
b67c463852a2: Inline CompilePhonemeData2 into espeak_ng_CompilePhonemeData.

Modified: src/libespeak-ng/compiledata.c


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
e0a598e9681e: Use variables for the phoneme source and destination paths.

Modified: src/libespeak-ng/compiledata.c


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
eb91c8dfceff: espeak_ng_CompilePhonemeDataPath: Allow the user to specify the source and destination paths for the phoneme data.

Modified: src/include/espeak-ng/espeak_ng.h
Modified: src/libespeak-ng/compiledata.c


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
e372d8434897: espeak-ng: add a argument to --compile-phonemes.

Modified: CHANGELOG.md
Modified: src/espeak-ng.c


Re: regarding assamese espeak

Valdis Vitolins
 

Hello, Leikhu!

eSpeakNG uses its own "programming language" to describe pronunciation as it is

described here:

https://github.com/espeak-ng/espeak-ng/blob/master/docs/add_language.md

https://github.com/espeak-ng/espeak-ng/blob/master/docs/intonation.md

To improve language you don't need programming knowledge, but it can help a lot,

because then you can do more tests independently at your computer as it is described at:

https://github.com/espeak-ng/espeak-ng

But if it is too hard for you, one of options you can do is to read just language files at:

https://github.com/espeak-ng/espeak-ng/blob/master/dictsource/as_list and

https://github.com/espeak-ng/espeak-ng/blob/master/dictsource/as_rules and test words at

http://odo.lv/Espeak?language=en&eslang=-vas and then submit comments for improvements as bugs or e-mails in this group.

Valdis


 


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by ValdisVitolins:
#199 Language analysis improvements

Language analysis and spelling decisions could be improved by introducing following new features: - [ ] extend verb follows/noun follows marks to more/arbitrary flags, which then can be used to make different pronunciation rules for homonyms - [ ] J statement as precondition, to allow choosing pronunciation from preceding word - [ ] possibility to go back to start of the rules and redo analysis again (I.e. not only after removing pre/suf-fixes. Could be performance drain, if used improperly.) - [ ] replace rule extended to replace not only characters, but group of characters, also probably replace using matching rules - [ ] _list extended to mark arbitrary defined word types by comparing only root part of the word (i.e. partial match without pre/suffixes) - [ ] output (prosody data) extended to mark syllables with more/arbitrary defined ways for different pronuciations (e.g. high/low pitch for Chinese etc.) - [ ] Fix issue #196 Word end mark _ doesn't work properly with ~ character group.


[espeak-ng:master] New Issue Created by ValdisVitolins:
#199 Language analysis improvements

Language analysis and spelling decisions could be improved by introducing following new features: - [ ] extend verb follows/noun follows marks to more/arbitrary flags, which then can be used to make different pronunciation rules for homonyms - [ ] J statement as precondition, to allow choosing pronunciation from preceding word - [ ] possibility to go back to start of the rules and redo analysis again (I.e. not only after removing pre/suf-fixes. Could be performance drain, if used improperly.) - [ ] replace rule extended to replace not only characters, but group of characters, also probably replace using matching rules - [ ] _list extended to mark arbitrary defined word types by comparing only root part of the word (i.e. partial match without pre/suffixes) - [ ] output (prosody data) extended to mark syllables with more/arbitrary defined ways for different pronuciations (e.g. high/low pitch for Chinese etc.) - [ ] Fix issue #196 Word end mark _ doesn't work properly with ~ character group.


[espeak-ng:master] New Issue Created by ValdisVitolins:
#199 Language analysis improvements

Language analysis and spelling decisions could be improved by introducing following new features: - [ ] extend verb follows/noun follows marks to more/arbitrary flags, which then can be used to make different pronunciation rules for homonyms - [ ] J statement as precondition, to allow choosing pronunciation from preceding word - [ ] possibility to go back to start of the rules and redo analysis again (I.e. not only after removing pre/suf-fixes. Could be performance drain, if used improperly.) - [x] replace rule extended to replace not only characters, but group of characters, also probably replace using matching rules - [ ] _list extended to mark arbitrary defined word types by comparing only root part of the word (i.e. partial match without pre/suffixes) - [ ] output (prosody data) extended to mark syllables with more/arbitrary defined ways for different pronuciations (e.g. high/low pitch for Chinese etc.) - [ ] Fix issue #196 Word end mark _ doesn't work properly with ~ character group.


regarding assamese espeak

Laishram Leingakpa
 

Dear group members
I would like to request Your valuable guidance to make improvement of
the Assamese espeak with an objective to configure the same with NVDA.
The already done basic version needs lot of refinement and
modification.
Moreover, Is it possible for a blind person to work on the same? If
so, what level of programming knowledge is needed?

Looking forward to your valuable responses.

--
Leikhu Laishram
Disability Activist
Mobile: +91 9954973243


Github push to espeak-ng:espeak-ng #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

1 New Commit:

[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
96e34565ed34: Use a default buffer length of 60mS to fix using MBROLA voices.

Modified: CHANGELOG.md
Modified: src/include/espeak-ng/speak_lib.h
Modified: src/libespeak-ng/speech.c


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

6 New Commits:

[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
059543ceb013: Fix for issue #195 Different non-Latin diacritized consonants are wrongly considered as one doubled

Modified: src/libespeak-ng/dictionary.c


[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
d9d7006d7aff: Comments about implementation for utf8_in2()

Modified: src/libespeak-ng/translate.c


[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
b10f6ea1a292: Warning deleted for fixed issue #195

Modified: docs/dictionary.md


[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
7ad277e39eff: Fixed typo

Modified: src/libespeak-ng/dictionary.c


[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
60396f3b5de2: Merge branch 'master' of https://github.com/espeak-ng/espeak-ng

Modified: .gitignore
Modified: android/src/com/reecedunn/espeak/CheckVoiceData.java
Modified: android/src/com/reecedunn/espeak/DownloadVoiceData.java


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
5b072d0fd629: Update the CHANGELOG.

Modified: CHANGELOG.md


[espeak-ng:master] Issue #195 Different non-Latin diacritized consonants are wrongly considered as one doubled closed by ValdisVitolins.


[espeak-ng:master] New Comment on Issue #195 Different non-Latin diacritized consonants are wrongly considered as one doubled
By rhdunn:

Thanks for the fix.


[espeak-ng:master] Label added to issue #195 Different non-Latin diacritized consonants are wrongly considered as one doubled by ValdisVitolins.


[espeak-ng:master] reported: Different non-Latin diacritized consonants are wrongly considered as one doubled #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Comment on Issue #195 Different non-Latin diacritized consonants are wrongly considered as one doubled
By ValdisVitolins:

It was hard weekend, but I managed to fix this.


Github push to espeak-ng:espeak-ng #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

2 New Commits:

[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
690bb8b7d24e: Ignore any present gradle wrapper.

Modified: .gitignore


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
d1c56e6120de: Android: Fix installing the voices -- use the new espeak-ng-data path.

Modified: android/src/com/reecedunn/espeak/CheckVoiceData.java
Modified: android/src/com/reecedunn/espeak/DownloadVoiceData.java


Github push to espeak-ng:espeak-ng #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

2 New Commits:

[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
58c655abaeeb: Use autoconf to check for endian.h.

Modified: configure.ac
Modified: src/include/compat/endian.h


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
2e84bbbcc812: Use autoconf to check for sys/endian.h.

Modified: configure.ac
Modified: src/include/compat/endian.h


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

2 New Commits:

[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
7cb79ea358c4: Don't use AC_FUNC_(M|RE)ALLOC as they cause 'Undefined reference to rpl_(m|re)alloc' on some systems.

Modified: configure.ac


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
39bbd4814435: README: Add a cross compilation section.

Modified: README.md


[espeak-ng:master] Issue #180 Undefined reference to rpl_malloc closed by midaszhou.


[espeak-ng:master] Label added to issue #180 Undefined reference to rpl_malloc by midaszhou.


[espeak-ng:master] Label added to issue #194 lkjsdf by sthibaul.


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Comment on Issue #180 Undefined reference to rpl_malloc
By midaszhou:

Thanks a lot! It works! For the 2nd step, I deleted 'AC_FUNC_MALLOC' and 'AC_FUNC_REALLOC' in configure.ac, then run autogen.sh and bulid.sh again. I edited build.sh for cross compilation like this: #!/bin/sh ./configure --build=i386-linux \ --host=mipsel-openwrt-linux \ --target=mips-linux

finially run make as ( make -B src/speak-ng src/espeak-ng ). Its tricky and exiting! image


[espeak-ng:master] New Comment on Issue #180 Undefined reference to rpl_malloc
By midaszhou:

Thanks a lot! It works! For the 2nd step, I deleted 'AC_FUNC_MALLOC' and 'AC_FUNC_REALLOC' in configure.ac, then run autogen.sh and bulid.sh again. I edited build.sh for cross compilation like this: #!/bin/sh ./configure --build=i386-linux \ --host=mipsel-openwrt-linux \ --target=mips-linux

finially run make as ( make -B src/speak-ng src/espeak-ng ). Its tricky and exiting! image


[espeak-ng:master] reported: Compiling on Arch Linux #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Comment on Issue #151 Compiling on Arch Linux
By xyproto:

Will try the above suggestions, thanks.

I am building in a clean chroot so there should be no root/sudo/permission issues. Will try building on another machine too.


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

1 New Commit:

[espeak-ng:master] By chrislm <llajta2012@...>:
fc7a94dcf14c: ES: minor fix for eI (vdipht) pronunciation

Modified: dictsource/es_list
Modified: dictsource/es_rules


[espeak-ng/espeak-ng] Pull request closed by rhdunn:

#198 ES: minor fix for eI (vdipht) pronunciation
Probably there was a mistake o'mine in last commit for Spanish... I edited es_rules / ".group e" to re-using the phoneme eI, such as in preview versions.


7 New Commits:

[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
91b723c99afe: Lowered nasaliness of Latvian voice

Modified: espeak-ng-data/voices/bat/lv


[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
47e1a2b43afd: Latvian language improvements

Modified: dictsource/lv_list
Modified: dictsource/lv_rules
Modified: espeak-ng-data/voices/bat/lv
Modified: phsource/ph_latvian


[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
fd30bdf026f8: More details about rule implementation and caveats

Modified: docs/dictionary.md


[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
640f0d8160ea: Typo fixes

Modified: src/libespeak-ng/voices.c
Modified: src/libespeak-ng/wavegen.c


[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
824e55b85592: Fixed link to bug

Modified: docs/dictionary.md


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
9d885c89fd4e: Merge remote-tracking branch 'ValdisVitolins/master'

Modified: dictsource/lv_list
Modified: dictsource/lv_rules
Modified: docs/dictionary.md
Modified: espeak-ng-data/voices/bat/lv
Modified: phsource/ph_latvian
Modified: src/libespeak-ng/voices.c
Modified: src/libespeak-ng/wavegen.c


[espeak-ng:master] By Reece H. Dunn <msclrhd@...>:
006bd842c39c: Fix indentation in wavegen.c.

Modified: src/libespeak-ng/wavegen.c


[espeak-ng/espeak-ng] Pull request closed by rhdunn:

#197 Some improvements for Latvian, code and documentation
Reece, can you mark issues related to language rules mark with appropriate tag? Thanks! Valdis

3941 - 3960 of 4111