[espeak-ng:master] New Comment on Issue #816 Wrong Identifiers
By gelodefaultbrain:

Hello I also tried this getting the IPA from Spanish Latin America

I tried this code data_ipa = phonemize(data,language="es-419",backend="espeak")

and it doesn't work it says unrecognized language code, do you might have happened to have a list of the correct codes or Identifiers? :D Thank you!

[espeak-ng:master] New Comment on Issue #816 Wrong Identifiers
By gelodefaultbrain:

Look :D when I tried "es-la" it worked! May I ask is this the same as "es-419", does "es-la" mean latin america? Thank you!

[espeak-ng:master] New Comment on Issue #816 Wrong Identifiers
By gelodefaultbrain:

Look :D when I tried "es-la" it worked! May I ask is this the same as "es-419", does "es-la" mean latin america? Thank you!

1 New Commit:

[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
ec7ebfb3c676: lv: add more unstressed words

Modified: dictsource/lv_list

[espeak-ng:master] New Comment on Issue #817 Broken import of multi-word terms from extra dictionary!?
By valdisvi:

Input words should be lowercase. If you want to mark that word should start or be entire in uppercase, you mark it with additional flags like $capital or $allcaps.

[espeak-ng:master] New Comment on Issue #817 Broken import of multi-word terms from extra dictionary!?
By repodiac:

Okay, thanks again. I close the issue then

[espeak-ng:master] Issue #817 Broken import of multi-word terms from extra dictionary!? closed by repodiac.

[espeak-ng:master] New Comment on Issue #817 Broken import of multi-word terms from extra dictionary!?
By repodiac:

Ok, I just checked out your suggestion: It works only, if I use lower-case so that the term is (common sense) !?! I first tried (Common Sense) which resulted in k'Omo:n z'Enz@ -- in literally pronounced German but not the correct English pronounciation!

Did I miss something or is only lowercase allowed with imports? Does this make sense overall, I mean you loose information on being substantive for a word for instance in German.

[espeak-ng:master] New Comment on Issue #817 Broken import of multi-word terms from extra dictionary!?
By repodiac:

Ok, I just checked out your suggestion: It works only, if I use lower-case so that the term is (common sense) !?! I first tried (Common Sense) which resulted in k'Omo:n z'Enz@ -- in literally pronounced German but not the correct English pronounciation!

Did I miss something or is only lowercase allowed with imports? Does this make sense overall, I mean you loose information on being substantive for a word for instance in German.

PS: I also checked the link you referred to above, it does not state that import dictionaries have to be lowercase...

[espeak-ng:master] New Comment on Issue #817 Broken import of multi-word terms from extra dictionary!?
By valdisvi:

Are you sure you have imported entry according to the rules? It should look like:

(common sense) ,kOm@n'||sEns

because otherwise second word sense is assumed to be pronunciation and everything else after that is ignored.

[espeak-ng:master] New Comment on Issue #817 Broken import of multi-word terms from extra dictionary!?
By valdisvi:

Probably yo have not imported entry according to the rules. It should look like:

(common sense) ,kOm@n'||sEns

because otherwise second word sense is assumed to be pronunciation and everything else after that is ignored.

[espeak-ng:master] New Comment on Issue #817 Broken import of multi-word terms from extra dictionary!?
By valdisvi:

Probably you have not imported entry according to the rules. It should look like:

(common sense) ,kOm@n'||sEns

because otherwise second word sense is assumed to be pronunciation and everything else after that is ignored.

[espeak-ng:master] New Comment on Issue #817 Broken import of multi-word terms from extra dictionary!?
By repodiac:

Oh great - thank you for the quick response! I'll look into it...

[espeak-ng:master] New Issue Created by repodiac:
#817 Broken import of multi-word terms from extra dictionary!?

Hi,

when working on a German loan word dictionary (https://github.com/espeak-ng/espeak-ng/issues/814), to be imported as de_extra into espeak-ng, I figured that often/always(?) terms with more than one word are either rejected or -- worse - imported but somewhat crippled.

An example: Importing the term common sense into the German de_extra dictionary results in S'ense z'Enz@

$ espeak-ng -v German "Common Sense" -x
S'ense z'Enz@

Is this as expected -- then: why? and how can I provide for a proper import? -- or is this likely a bug? Thanks for any help!

[espeak-ng:master] New Issue Created by repodiac:
#817 Broken import of multi-word terms from extra dictionary!?

Hi,

when working on a German loan word dictionary (https://github.com/espeak-ng/espeak-ng/issues/814), to be imported as de_extra into espeak-ng, I figured that often/always(?) terms with more than one word are either rejected or -- worse - imported but somewhat crippled.

An example: Importing the term common sense as entry ,kOm@n'sEns into the German de_extra dictionary results in S'ense z'Enz@

$ espeak-ng -v German "Common Sense" -x
S'ense z'Enz@

Is this as expected -- then: why? and how can I provide for a proper import? -- or is this likely a bug? Thanks for any help!

[espeak-ng:master] New Comment on Issue #816 Wrong Identifiers
By gelodefaultbrain:

Hello! Thank you so much for your reply I am using this to get the IPA of words from different languages :D Here is the code where I use the phonemizer library :D

The last time I tried for the language french and on the documentation it was said to be "fr" then I discovered on your other docs using the espeak ng -v [LANG] I think that's on the terminal right? that it is "fr-fr" for french so I got it! :D

Now I am trying right now for the West Germanic English Received Pronunciations and I tried the en-gb-x-rp and as you can see on the previous images that it doesn't recognize it, and I think I am just inputting the wrong specifier for it so may I ask what would be the right specifier for it? Thank you! this library is really awesome and helps a lot :D

[espeak-ng:master] New Comment on Issue #816 Wrong Identifiers
By gelodefaultbrain:

data_ipa = phonemize(data,language="en-gb-x-rp",backend="espeak")

This does not work and it doesn't recognize the language parameter value which is en-gb-x-rp I am just thinking that I am putting the wrong specifier just like what happened with French cause what I did is put "fr" and I soon discovered that it should be "fr-fr" so maybe yeah I am just putting the wrong specifier :D would you happen to have a list of it? :D Thank you!

[espeak-ng:master] New Comment on Issue #816 Wrong Identifiers
By gelodefaultbrain:

By the way I am not using any of the commands using the terminal, tho I've tried it :D as of now I am using this with python and it's really helpful

[espeak-ng:master] New Comment on Issue #816 Wrong Identifiers
By rhdunn:

The identifiers marked are the BCP 47 language tags for the supported languages. Because fr has a default country code of FR, you can omit that from the tag as is done here. Also, the dialects use the private tags (-x-[tag]) from https://raw.githubusercontent.com/espeak-ng/bcp47-data/master/bcp47-extensions. Also note that BCP 47 tags are case insensitive.

The language file for en-gb-x-rp is lang/gmw/en-GB-x-rp, so maybe that issue is with a case sensitive check/match? However, note that the file contains the following:

language en-gb-x-rp
language en-gb  4
language en 5

so using the (espeak_/espeakng)SetVoiceByName/SetVoiceByProperties APIs should be matching the language correctly.

The French (France) voice is located in lang/roa/fr, so that is correct. Note that that file defines the following languages:

language fr-fr
language fr

so should accept either fr or fr-fr.

What program/python binding are you using? Do you know how that python binding is checking for the language? Do you get the same issue using espeak-ng -v [LANG]?

[espeak-ng:master] New Comment on Issue #816 Wrong Identifiers
By gelodefaultbrain:

"en-gb" works but that "en-gb-x-rp" does not work

[espeak-ng:master] New Comment on Issue #816 Wrong Identifiers
By gelodefaultbrain:

Also on your .md it says "fr" and it should be "fr-fr" cause in python and on the documentation you have you guys indicated it "fr-fr" that's where I learned that it is not "fr". It's kinda confusing and it might help others too. Thank you!

[espeak-ng:master] New Issue Created by gelodefaultbrain:
#816 Wrong Identifiers

Hello! I would like to share my experience with this, this library is great and it helped me a lot :D

Just want to add up that the identifier for French is not "fr" I think ti should be indicated there "fr-fr" and I am thinking that this could apply to other languages as well. In this British right there where I boxed it, it is not identified in python , I am thinking that it's the wrong identifier again. Can you guys double check it :D Thank you!

[espeak-ng:master] New Comment on Issue #7 Reimplement the SAPI bindings.
By amirsol81:

Folks, may I humbly know why no SAPI 5 installer exists or hasn't been released for this project? It will open lots of opportunities on Windows in combination with apps like JAWS. The old eSpeak SAPI 5 is quite crash-prone on Windows.

7 New Commits:

[espeak-ng:master] By Juho Hiltunen <jaacoppi@...>:
50f58168e100: code cleanup: remove unused *constcharptr and MakeWawe2()

Modified: src/libespeak-ng/synthesize.h
Modified: src/libespeak-ng/translate.h

[espeak-ng:master] By Juho Hiltunen <jaacoppi@...>:
75d66c89d035: code cleanup: Move self_closing checks to ProcessSsmlTag()

This is a bit slower since we don't pass n_xml_buf as an argument but
rather get it with a call to wcslen. It is much cleaner though, since
the name ProcessSsmlTag() implies that all processing should be done
there.

Modified: src/libespeak-ng/readclause.c
Modified: src/libespeak-ng/ssml.c
Modified: src/libespeak-ng/ssml.h

[espeak-ng:master] By Juho Hiltunen <jaacoppi@...>:
34657e7ea4fb: code cleanup: move check for SSML comments and declarations to
ProcessSsmlTag()

Note the line in readclause.c:
if ((c2 == '/') || iswalpha(c2) || c2 == '!' || c2 == '?') {

It might be enough to pass everything to ProcessSsmlTag. What are the
cases that are skipped because of this?

Modified: src/libespeak-ng/readclause.c
Modified: src/libespeak-ng/ssml.c

[espeak-ng:master] By Juho Hiltunen <jaacoppi@...>:
959bf26b6b1a: Add a test for XML/SSML character and entity references

Added: tests/ssml/references.expected
Added: tests/ssml/references.ssml2
Modified: tests/ssml.test

[espeak-ng:master] By Juho Hiltunen <jaacoppi@...>:
54d93cf2b4bc: code cleanup: move ssml reference handling logic to a new function ParseSsmlReference()

It's unclear why c2 needs to be set after an entity reference.

Modified: src/libespeak-ng/readclause.c
Modified: src/libespeak-ng/ssml.c
Modified: src/libespeak-ng/ssml.h

[espeak-ng:master] By Juho Hiltunen <jaacoppi@...>:
c9dee003bf62: better explanation and nicer formatting for the ssml reference test.

Modified: tests/ssml/references.ssml2

[espeak-ng:master] By Valdis Vitolins <valdis.vitolins@...>:
5b3da5950aea: Merge pull request #813

Added: tests/ssml/references.expected
Added: tests/ssml/references.ssml2
Modified: src/libespeak-ng/readclause.c
Modified: src/libespeak-ng/ssml.c
Modified: src/libespeak-ng/ssml.h
Modified: src/libespeak-ng/synthesize.h
Modified: src/libespeak-ng/translate.h
Modified: tests/ssml.test

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

#813 SSML cleanup

Move SSML related logic from ReadClause to ProcessSsmlTag and to a new function ParseSsmlReference().

Contributes to #369 and should make locating SSML bugs easier.

[espeak-ng:master] New Comment on Pull Request #813 SSML cleanup
By jaacoppi:

Thanks, live and learn.. I was only testing with an ssml string before refactoring, but wrote the committed test with comments afterwards.

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

#813 SSML cleanup

Move SSML related logic from ReadClause to ProcessSsmlTag and to a new function ParseSsmlReference().

Contributes to #369 and should make locating SSML bugs easier.

[espeak-ng:master] New Comment on Pull Request #813 SSML cleanup
By jaacoppi:

I didn't even notice the test failing as it was added later.

Running this on current master branch: espeak-ng --punct -x -m -f input.txt with input.txt being a simple one liner <!-- Commenting --><speak>&apos;</speak> Produces: gr'eIt@D,an_: t'Ik.

That is clearly wrong. It seems that the current way of handling combination of comments and entity references is broken somehow, and 34657e7 fixes it. My aim was to change nothing, just reorganize the code, so I certainly don't know why this is happening.

The extra spaces in references.ssml2 are there by accident. I can add a commit for reformatting. Is this good?

<speak>&lt; &gt; &amp; &apos; &quot;</speak>
<speak>B &#66;</speak>
<speak>z &#x7A;</speak>

[espeak-ng:master] New Comment on Pull Request #813 SSML cleanup
By valdisvi:

I think, it is old problem which will be fixed, please add commit to make xml formatted nicer, and I'll merge it. In future, it is better to start by adding test before refactoring. That will allow to ensure, that it either didn't come worse, or it actually fixes some problem which was not found before.

[espeak-ng:master] New Comment on Pull Request #813 SSML cleanup
By valdisvi:

I think, it is old problem which will be fixed. Please add commit to make xml formatted nicer, and I'll merge it. In future, it is better to start by adding test before refactoring. That will allow to ensure, that it either didn't come worse, or it actually fixes some problem which was not found before.

[espeak-ng:master] New Comment on Pull Request #813 SSML cleanup
By jaacoppi:

I didn't even notice the test failing as it was added later.

Running this on current master branch: espeak-ng --punct -x -m -f input.txt with input.txt being a simple one liner <!-- Commenting --><speak>&apos;</speak> Produces: gr'eIt@D,an_: t'Ik.

That is clearly wrong. It seems that the current way of handling combination of comments and entity references is broken somehow, and 34657e7 fixes it. My aim was to change nothing, just reorganize the code, so I certainly don't know why this is happening.

The extra spaces in references.ssml2 are there by accident. I can add a commit for reformatting. Is this good? < > & ' " B B z z

[espeak-ng:master] New Comment on Pull Request #813 SSML cleanup
By valdisvi:

To check for regressions, I first cherry-picked only commit 959bf26 and run make check with everything else the same. I got error:

testing tests/ssml/references.ssml2
1c1
< l'EsDan_: gr'eIt@D,an_: 'amp@s,and t'Ik_: kw'oUts
---
> gr'eIt@D,an_: l'EsDan_: gr'eIt@D,an_: 'amp@s,and t'Ik_: kw'oUts
make: *** [Makefile:2654: tests/ssml.check] Error 1

Does it means, that current implementation actually is broken and doesn't work as expected? Why in tests/ssml/references.ssml2

<speak> &lt; &gt; &amp; &apos; &quot; </speak>
<speak> B &#66;</speak>
<speak>z &#x7A;</speak>

with all commits applied it works with spaces removed.

[espeak-ng:master] New Comment on Pull Request #813 SSML cleanup
By valdisvi:

To check for regressions, I first cherry-picked only commit 959bf26 and run make check with everything else the same. I got error:

testing tests/ssml/references.ssml2
1c1
< l'EsDan_: gr'eIt@D,an_: 'amp@s,and t'Ik_: kw'oUts
---
> gr'eIt@D,an_: l'EsDan_: gr'eIt@D,an_: 'amp@s,and t'Ik_: kw'oUts
make: *** [Makefile:2654: tests/ssml.check] Error 1

Does it means, that current implementation actually is broken and doesn't work as expected? Why in tests/ssml/references.ssml2 are unneded spaces?

<speak> &lt; &gt; &amp; &apos; &quot; </speak>
<speak> B &#66;</speak>
<speak>z &#x7A;</speak>

with all commits applied it works with these spaces removed.

[espeak-ng:master] New Comment on Pull Request #813 SSML cleanup
By valdisvi:

To check for regressions, I first cherry-picked only commit 959bf26 and run make check with everything else the same. I got error:

testing tests/ssml/references.ssml2
1c1
< l'EsDan_: gr'eIt@D,an_: 'amp@s,and t'Ik_: kw'oUts
---
> gr'eIt@D,an_: l'EsDan_: gr'eIt@D,an_: 'amp@s,and t'Ik_: kw'oUts
make: *** [Makefile:2654: tests/ssml.check] Error 1

Does it means, that current implementation actually is broken and doesn't work as expected? Why in tests/ssml/references.ssml2 are unneded spaces?

<speak> &lt; &gt; &amp; &apos; &quot; </speak>
<speak> B &#66;</speak>
<speak>z &#x7A;</speak>

with all commits applied it works with these spaces removed.

[espeak-ng:master] New Comment on Issue #737 SSML input changes to default voice mid-sentence
By jaacoppi:

One options is to parse the XML as a document.

That's the long term plan in #369. I've been restructuring SSML support for a while now. However, the code is filled with global variables so restructuring is not easy.