Date   

Updates to Github #github

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

[espeak-ng:master] New Comment on Issue #1016 new ESpeak variants wider variety of speaking styles
By joshknnd1982:

It wouldn’t mimmick the actual voice, just the intonations of that voice.

Sent from Mail for Windows

From: Valdis Vitolins Sent: Wednesday, September 15, 2021 15:46 To: espeak-ng/espeak-ng Cc: joshknnd1982; Author Subject: Re: [espeak-ng/espeak-ng] new ESpeak variants wider variety ofspeaking styles (#1016)

Intonation of eSpeak NG voices is described in Intonation page. I don't know how specifics for mentioned tools, but due to technological differences, possibility to mimic voice of one tool as voice of another is quite improbable. Just read basics of e.g. eSpeakNG, MBROLA and Klatt. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.


[espeak-ng:master] New Comment on Issue #1016 new ESpeak variants wider variety of speaking styles
By joshknnd1982:

Ok, so how do I run the:

espeak-ng --compile-intonations

how do I run that command? I am using windows10 with a compiled Espeak-ng as part of NVDA screen reader.

Sent from Mail for Windows

From: Valdis Vitolins Sent: Wednesday, September 15, 2021 15:46 To: espeak-ng/espeak-ng Cc: joshknnd1982; Author Subject: Re: [espeak-ng/espeak-ng] new ESpeak variants wider variety ofspeaking styles (#1016)

Intonation of eSpeak NG voices is described in Intonation page. I don't know how specifics for mentioned tools, but due to technological differences, possibility to mimic voice of one tool as voice of another is quite improbable. Just read basics of e.g. eSpeakNG, MBROLA and Klatt. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.


[espeak-ng:master] reported: Add luxembourgish #github

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

[espeak-ng:master] New Comment on Issue #1013 Add luxembourgish
By valdisvi:

Adding artificial languages is actually easier than adding native ones, because they are simpler and more consistent.


Updates to Github #github

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

[espeak-ng:master] New Comment on Issue #1016 new ESpeak variants wider variety of speaking styles
By valdisvi:

Intonation of eSpeak NG voices is described in Intonation page. I don't know how specifics for mentioned tools, but due to technological differences, possibility to mimic voice of one tool as voice of another is quite improbable. Just read basics of e.g. eSpeakNG, MBROLA and Klatt.


[espeak-ng:master] New Comment on Issue #1016 new ESpeak variants wider variety of speaking styles
By valdisvi:

Intonation of eSpeak NG voices is described in Intonation page. I don't know specifics of mentioned tools, but due to technological differences, possibility to mimic voice of one tool as voice of another is quite improbable. Just read basics of e.g. eSpeakNG, MBROLA and Klatt.


[espeak-ng:master] New Comment on Issue #1016 new ESpeak variants wider variety of speaking styles
By valdisvi:

Intonation of eSpeak NG voices is described in Intonation page. I don't know specifics of mentioned tools, but due to technological differences, possibility to mimic voice of one tool as voice of another is quite improbable. Just read basics of e.g. eSpeakNG, MBROLA and Klatt.


[espeak-ng:master] New Comment on Issue #1016 new ESpeak variants wider variety of speaking styles
By valdisvi:

Intonation of eSpeak NG voices is described in Intonation page. I don't know specifics of mentioned tools, but due to technological differences, possibility to mimic voice of one tool as voice of another is quite improbable. Just read basics of e.g. eSpeakNG, MBROLA and Klatt and then about HTS, TensorFlow-WaveNet and many other ones.


[espeak-ng:master] reported: MBROLA support Android #github

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

[espeak-ng:master] New Comment on Issue #1010 MBROLA support Android
By valdisvi:

MBROLA is separate speech synthesis project and, as far as I know, it is not ported to Android. MBROLA needs data files for speech synthesis, which are available in MBROLA-voices project.


[espeak-ng:master] reported: A TTS is too chatty and "smart" and at most situations a builtin converter from acronims - abriviations to words is not accurate. #github

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

[espeak-ng:master] New Comment on Issue #1014 A TTS is too chatty and "smart" and at most situations a builtin converter from acronims - abriviations to words is not accurate.
By valdisvi:

I can't reproduce described behavior in development version of eSpeak NG for Slovenian:

espeak-ng -x -vsl "vi x64 10 cm"
S'e:st 'iks St'i:RiinS'e:zddEsEd dEs'e:t ts,@m'@

For English only vi is pronounced as "roman six":

espeak-ng -x -ven "vi x64 10 cm"
r,oUm@n_ s'Iks 'Eks s'Iksti f'o@ t'En s,i:;'Em


Pull Request Updated #github

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

[espeak-ng/espeak-ng] Pull request updated by alex19EP:

#960 separate the android project from the make system.

WIP

I want to make the build of android application independent of make and autoconf.

this will simplefy building the application on non-linux platforms and improve project usability with android-studio.


[espeak-ng:master] new issue: new ESpeak variants wider variety of speaking styles #github

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

[espeak-ng:master] New Issue Created by joshknnd1982:
#1016 new ESpeak variants wider variety of speaking styles

Could someone in charge of voice variants add new tunes statements to the ESpeak intonation file to both emulate other tts engines speaking styles and also make new and different voices with different speaking styles, based on additional tunes statements? I would like to have an ESpeak voice that speaks with the inflections of doubleTalk or trippleTalk speech, another that speaks with Keynote gold intonation style, and perhaps another with a similar intonation style to infovox230 or dolphin orpheus. Perhaps other ESpeak voices that sing the text in some way or have other speaking styles based on new tunes statements. with espeak sapi5 you would open the intonations source file in notepad, add new tunes statements, recompile, then make a new voice variant file with the new tunes statements added at the end such as, "tunes s8, c8, e8, q8".


[espeak-ng:master] new issue: `make` breaks on spaces in the path #github

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

[espeak-ng:master] New Issue Created by TheDutchMC:
#1015 `make` breaks on spaces in the path

When working in WSL, it is common for me to work directly from the Windows Downloads directory e.g. My username contains a space, this breaks the makefile:

ESPEAK_DATA_PATH=/mnt/c/Users/Tobias de Bruijn/Downloads/espeak-ng-1.50/espeak-ng src/espeak-ng --compile-intonations && \
        ESPEAK_DATA_PATH=/mnt/c/Users/Tobias de Bruijn/Downloads/espeak-ng-1.50/espeak-ng src/espeak-ng --compile-phonemes && \
        touch phsource/phonemes.stamp
/bin/bash: de: command not found
make[1]: *** [Makefile:2781: phsource/phonemes.stamp] Error 127
make[1]: Leaving directory '/mnt/c/Users/Tobias de Bruijn/Downloads/espeak-ng-1.50/espeak-ng'
make: *** [Makefile:751: all] Error 2


Updates to Github #github

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

[espeak-ng:master] New Comment on Issue #1014 A TTS is too chatty and "smart" and at most situations a builtin converter from acronims - abriviations to words is not accurate.
By gregjozk:

It is general observation, but seen mostly during english and slovenian use, which have been used daily.

2021-09-14 11:15 GMT+02.00, jaacoppi @.***>: > All TTS software pronounce words based on assumptions about the language and > context of the sentence. The modern solution would be to pass meta > information about the sentence to the TTS. One such solution is part of > speech tagging (POS). > > Since the origins of espeak NG are in 1995, these modern solutions are not > used. Instead, all choices are rule based. It would be great to give user > more configuration options to choose the preferred behavior. That is a work > in progress with LoadVoice() and LoadConfig() functions. Unfortunately the > progress is slow because the original code wasn't designed to support all > these languages and situations. > > Currently, the only option for the user is to edit the config files and > rebuild the code. That's probably too difficult for most users. > > > -- > You are receiving this because you authored the thread. > Reply to this email directly or view it on GitHub: > https://github.com/espeak-ng/espeak-ng/issues/1014#issuecomment-918968335


[espeak-ng:master] New Comment on Issue #1014 A TTS is too chatty and "smart" and at most situations a builtin converter from acronims - abriviations to words is not accurate.
By valdisvi:

Because of reasons @gregjozk mentioned, particularly for Latvian, there are almost no expansion of abbreviations except very few unique ones, because they are different in different domains. Therefore, for most of cases, there is no acute need to adjust behavior dynamically, unneeded expansions of abbreviations just need to be removed from rules.


Updates to Github #github

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

[espeak-ng:master] New Comment on Issue #1014 A TTS is too chatty and "smart" and at most situations a builtin converter from acronims - abriviations to words is not accurate.
By valdisvi:

About which language are you talking about?


[espeak-ng:master] New Comment on Issue #1014 A TTS is too chatty and "smart" and at most situations a builtin converter from acronims - abriviations to words is not accurate.
By jaacoppi:

All TTS software pronounce words based on assumptions about the language and context of the sentence. The modern solution would be to pass meta information about the sentence to the TTS. One such solution is part of speech tagging (POS).

Since the origins of espeak NG are in 1995, these modern solutions are not used. Instead, all choices are rule based. It would be great to give user more configuration options to choose the preferred behavior. That is a work in progress with LoadVoice() and LoadConfig() functions. Unfortunately the progress is slow because the original code wasn't designed to support all these languages and situations.

Currently, the only option for the user is to edit the config files and rebuild the code. That's probably too difficult for most users.


[espeak-ng:master] new issue: Add luxembourgish #github

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

[espeak-ng:master] New Issue Created by hozosch:
#1013 Add luxembourgish

Ok, I know, another of those add this or that language things and there are plenty of other difficult issues for you to deal with, not just concerning this project. Really, if I had the time and patience, I'd add the language myself and create a pull request instead. Saying that, so far, all I did is improve existing languages, and adding a completely new language feels like a much more intimidating task for me. Sure, I could read all the guidelines and stuff, but here go time and patience again. When it comes to the language itself, one argument not to add this language could be that German and French, which already exist in ESpeak, are spoken and / or understood by luxembourgers as well, so why add this language? Well, ESpeak has all kinds of minority languages, artificial languages, and whatnot. So I don't see the point in not adding it when I think about it like that. Besides, there's luxembourgish as a display language in Windows, and wouldn't a blind person in luxembourg who uses luxembourgish on a regular basis want to have a TTS being able to speak in that language to him / her? NVDA doesn't speak it yet, but that should be fairly easy to do. All I ask is that you create an initial basis on which I'll provide patches later on if needed. You may use this as your base: https://en.wikipedia.org/wiki/Luxembourgish_phonology Thanks very, very much in advance for considering this!


[espeak-ng:master] new issue: Add luxembourgish #github

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

[espeak-ng:master] New Issue Created by hozosch:
#1013 Add luxembourgish

Ok, I know, another of those add this or that language things and there are plenty of other difficult issues for you to deal with, not just concerning this project. Really, if I had the time and patience, I'd add the language myself and create a pull request instead. Saying that, so far, all I did is improve existing languages, and adding a complete new language feels like a much more intimidating task for me. Sure, I could read all the guidelines and stuff, but here go time and patience again. When it comes to the language itself, one argument not to add this language could be that German and French, which already exist in ESpeak, are spoken and / or understood by luxembourgers as well, so why add this language? Well, ESpeak has all kinds of minority languages, artificial languages, and whatnot. So I don't see the point in not adding it when I think about it like that. Besides, there's luxembourgish as a display language in Windows, and wouldn't a blind person in luxembourg who uses luxembourgish on a regular basis want to have a TTS being able to speak in that language to him / her? NVDA doesn't speak it yet, but that should be fairly easy to do. All I ask is that you create an initial basis on which I'll provide patches later on if needed. You may use this as your base: https://en.wikipedia.org/wiki/Luxembourgish_phonology Thanks very, very much in advance for considering this!


[espeak-ng:master] new issue: A TTS is too chatty and "smart" and at most situations a builtin converter from acronims - abriviations to words is not accurate. #github

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

[espeak-ng:master] New Issue Created by gregjozk:
#1014 A TTS is too chatty and "smart" and at most situations a builtin converter from acronims - abriviations to words is not accurate.

Hello,

I'm raising this issue, due to on the NVDA's site on Github there are more and more issues, which are firstly linked to TTS and a default TTS in NVDA is espeak NG or Onecore, which seems to inherit most code from espeak NG due to same mistakes that it is doing.

here is a problem: a synth treats both acronims and abriviations the same so "cm" is always "centimeter" but in most cases it is not correct. there are many more cases, where this behaviour is incorrect. e.g.: in slavic languages "you" in plural is "vi", but espeak it treats always as "roman 6".

also test, how is spelled "x64". in most cases it is pronounced as "ten sixty four".

possible solution: do not convert abriviations to words, because in different situations they have different meanings an leave tts to spell it or speak it as usual word if it is written with lowercase included.

thanks.

regards, Jožef


[espeak-ng:master] new issue: Add luxembourgish #github

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

[espeak-ng:master] New Issue Created by hozosch:
#1013 Add luxembourgish

Ok, I know, another of those add this or that language things and there are plenty of other difficult issues for you to deal with, not just concerning this projekt. Really, if I had the time and patience, I'd add the language myself and create a pull request instead. Saying that, so far, all I did is improve existing languages, and adding a complete new language feels like a much more intimidating task for me. Sure, I could read all the guidelines and stuff, but here go time and patience again. When it comes to the language itself, one argument not to add this language could be that German and French, which already exist in ESpeak, are spoken and / or understood by luxembourgers as well, so why add this language? Well, ESpeak has all kinds of minority languages, artificial languages, and whatnot. So I don't see the point in not adding it when I think about it like that. Besides, there's luxembourgish as a display language in Windows, and wouldn't a blind person in luxembourg who uses luxembourgish on a regular basis want to have a TTS being able to speak in that language to him / her? NVDA doesn't speak it yet, but that should be fairly easy to do. All I ask is that you create an initial basis on which I'll provide patches later on if needed. You may use this as your base: https://en.wikipedia.org/wiki/Luxembourgish_phonology Thanks very, very much in advance for considering this!


[espeak-ng:master] reported: some languages in espeakNG are broken #github

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

[espeak-ng:master] New Comment on Issue #1009 some languages in espeakNG are broken
By valdisvi:

I made changes for R and ain sound, hopefully with improvement. General rule for calculation of stress is not easy, because rules of stress for Arabic language are applied for pronounced sounds (phonemes), that is when pronunciation is already decided. But eSpeak allows to calculate stress regarding to written characters during handling of the text. And it is not straightforward even with added diacritics, because some characters may be pronounced either as vowel or consonant (e.g. ا و ي). Therefore it will need a lot of particular rules and it never be consistent enough. For proper solution eSpeak NG needs more additional rules for phoneme conditions, where stress could be changed depending on on type of following phonemes till to the end of the word. We discussed about that and integration with Mishkal in F123 Consulting project, but their funding was suspended. As a work-in-progress there is Arabic branch in my fork of the project, where eSpeak calls Mishkal for diacritization before pronunciation. But it was not clean solution, only proof of concept, therefore it was not merged into main project. Pronunciation of Arabic is too complicated to get good results only with occasional volunteers (you may do it, but then it should be your hobby of your lifetime). If you want to improve it faster, I suggest you to get financial grant to buy time of project maintainers. The same is true for Persian, Hebrew and other abjad languages.


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

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

1 New Commit:

[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
02783150cd01: ar: issue #1009: improve sound of Ain

Modified: phsource/ph_arabic
Modified: tests/language-phonemes.test


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

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

1 New Commit:

[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
aa765c76807f: ar: issue #1009: change r to trilling R

Modified: dictsource/ar_list
Modified: dictsource/ar_rules
Modified: tests/language-pronunciation.test


[espeak-ng:master] reported: some languages in espeakNG are broken #github

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

[espeak-ng:master] New Comment on Issue #1009 some languages in espeakNG are broken
By king-dahmanus:

good luck sir. And don't forget to change the raa letter. Because as i said, replacing r with capital R fixes the issue

On Mon, 13 Sept 2021 at 15:00, jaacoppi @.***> wrote:

Yeah, I missed your explanation for the latter rule. Most rules can be implemented in espeak NG. We'll just need a definition that is precise enough. I'll start looking at this: https://en.wikipedia.org/wiki/Arabic_phonology#Word_stress

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/espeak-ng/espeak-ng/issues/1009#issuecomment-918220908, or unsubscribe https://github.com/notifications/unsubscribe-auth/AT2FKJXG4I73TURXJVA4QL3UBX7WXANCNFSM5DP376CA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.


[espeak-ng:master] reported: some languages in espeakNG are broken #github

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

[espeak-ng:master] New Comment on Issue #1009 some languages in espeakNG are broken
By jaacoppi:

Yeah, I missed your explanation for the latter rule. Most rules can be implemented in espeak NG. We'll just need a definition that is precise enough. I'll start looking at this: https://en.wikipedia.org/wiki/Arabic_phonology#Word_stress


[espeak-ng:master] reported: some languages in espeakNG are broken #github

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

[espeak-ng:master] New Comment on Issue #1009 some languages in espeakNG are broken
By king-dahmanus:

never mind the example, i already am giving feadback on it. in the almut:'aqi:n exapmle that you gave me, it's wrong. It should be [[?almut:aq'i:n]]. Can't you somehow make espeak stress a specific letter whenever a specific letter is detected after it in a specific pattern? Or is that one of the disadvantages of rule based synthesis?/espeak's way of working?

281 - 300 of 4904