search

nvaccess / nvda

NVDA, the free and open source Screen Reader for Microsoft Windows
https://www.nvaccess.org/
Other
2.11k stars 637 forks source link

request to add phonetics to Indian languages #4410

Closed nvaccessAuto closed 9 years ago

nvaccessAuto commented 10 years ago

Reported by sumandogra on 2014-08-23 08:51 Requirement of addition of feature of phonetics to various Indian languages in NVDA similar to English Right now, on pressing NUMPAD 2 for various Indian languages no phonetic is available. Request is to add this feature so that it will be helpful for various Indian languages such as Hindi.

nvaccessAuto commented 9 years ago

Attachment bn_characterDescriptions.dic added by siddhartha_iitd on 2014-12-12 08:07 Description: Bengali Character Descriptions

nvaccessAuto commented 9 years ago

Attachment gu_characterDescriptions.dic added by siddhartha_iitd on 2014-12-12 08:08 Description: Gujarati Character Descriptions

nvaccessAuto commented 9 years ago

Attachment kok_characterDescriptions.dic added by siddhartha_iitd on 2014-12-12 08:08 Description: Konkani Character Descriptions

nvaccessAuto commented 9 years ago

Attachment ml_characterDescriptions.dic added by siddhartha_iitd on 2014-12-12 08:08 Description: Malayalam Character Descriptions

nvaccessAuto commented 9 years ago

Attachment pa_Character Descriptions.txt added by siddhartha_iitd on 2014-12-12 08:09 Description: Punjabi Character Descriptions

nvaccessAuto commented 9 years ago

Attachment mni_characterDescriptions.dic added by siddhartha_iitd on 2014-12-18 10:11 Description: Manipuri Character Descriptions

nvaccessAuto commented 9 years ago

Attachment te_characterDescriptions.dic added by siddhartha_iitd on 2014-12-18 10:14 Description: Telugu Character Descriptions

nvaccessAuto commented 9 years ago

Attachment as_characterDescriptions.dic added by siddhartha_iitd on 2015-01-05 07:16 Description: Assamese Character Descriptions

nvaccessAuto commented 10 years ago

Comment 1 by nvdakor on 2014-08-23 09:07 Hi, This requires character descriptions file for Indic languages. I'll change the component in hopes of getting input from translators.

nvaccessAuto commented 10 years ago

Comment 2 by blindbhavya on 2014-08-25 10:31 Hi. Mr. Him Prasad Gautam and I are the Hindi maintainers of NVDA. We started this very recently so have only almost completed translation and correction in translation of NVDA interface messages. After completion of correction we will begin translating other files such as Character Descriptions. In terms of Indian languages, only Hindi and Tamil are have been translated and are maintained.

nvaccessAuto commented 9 years ago

Comment 4 by siddhartha_iitd on 2014-12-01 07:20 characterDescriptions.dic files for Indian languages - Hindi, Bengali, Punjabi, Malayalam, Gujarati, Konkani - are available at the bitbucket repository mentioned below. We are also trying to get such description files for other Indian Languages.

Please check the branch in_t4410 at the below mentioned url: https://manish_agrawal@bitbucket.org/manish_agrawal/nvda.git

nvaccessAuto commented 9 years ago

Comment 5 by jteh on 2014-12-01 09:57 Siddhartha and Bhavya, are you collaborating here or are these separate efforts? Because we already have Hindi translators on board, they need to be managing/approving any changes. Also, because this is a full translation, any work needs to be submitted using the srt system (separate svn repository).

For the other languages, I'm not quite sure how to manage this. Mesar, there doesn't seem to be any interest in translating anything but character descriptions (and maybe symbols?) for other Indian languages. Do you think it makes sense for us to just commit the files directly to NVDA Git or do you think they should be managed by srt even so?

nvaccessAuto commented 9 years ago

Comment 6 by blindbhavya on 2014-12-01 12:37 Hi, Currently, these are separate efforts. Siddhartha, would you like to join the Hindi translation team, there is certainly a need for more maintainers for Hindi (for untranslated parts of User Guie, add-ons, and most parts are left).

nvaccessAuto commented 9 years ago

Comment 8 by siddhartha_iitd on 2014-12-12 08:13 The character description files for Indian languages - Bengali, Gujarati, Konkani, Malayalam, Punjabi have been attached at the ticket. Moreover, these files are also available at branch in_t4410 at following url: https://manish_agrawal@bitbucket.org/manish_agrawal/nvda.git

nvaccessAuto commented 9 years ago

Comment 9 by chrislm on 2014-12-12 09:27 In Hindi characters the symbol: ् '\u094d' is not pronunced by eSpeak synthesizer... Could we send a request to Espeak project?

nvaccessAuto commented 9 years ago

Comment 10 by siddhartha_iitd (in reply to comment 9) on 2014-12-12 09:49 Replying to chrislm:

In Hindi characters the symbol:

् '\u094d'

is not pronunced by eSpeak synthesizer...

Could we send a request to Espeak project?

This is similar to the issue 3805, which says some Unicode characters are not spoken by NVDA. To rectify this issue, I added this character along with its description in symbols.dic file with the level as either some or most.

nvaccessAuto commented 9 years ago

Comment 11 by chrislm (in reply to comment 9) on 2014-12-12 10:18 Replying to chrislm:

In Hindi characters the symbol:

् '\u094d'

is not pronunced by eSpeak synthesizer...

Sorry, I saw jus now, eSpeak 1.48.13 pronounces all those characters. I don't think this is a issue described in 3805, I tried eSpeak 1.48.13 in nvda master source and the character is pronounced by eSpeak. Also, eSpeak TestLatest SAPI5 recognizes this character correctly. Very thanks.

nvaccessAuto commented 9 years ago

Comment 12 by dhankuta on 2014-12-14 07:03 Hi, Currently being the Hindi maintainer, The characterDescription.dic file for Hindi is already committed by me. In case of Punjabi, Mr. Dinesh Mital punjabimaster259@gmail.com is the translator of Punjabi. Better to contact and provide the file to him and get committed. In case of the Halanta sign: Basically, a halanta suppresses the inherited vowel of the preceding consonant. It has no its own pronunciation if followed to a consonant. Though, grammatically, its presence in standalone state or if not preceded by a consonant, is wrong; however, if prevailed, its name must be pronounced. But due to less attention, of the Hindi espeak dictionery writer, its pronunciation was absent in previous versions. Now, after my suggestion, it is included.

nvaccessAuto commented 9 years ago

Comment 13 by MHameed (in reply to comment 4) on 2015-01-10 10:26 Replying to jteh:

Mesar, there doesn't seem to be any interest in translating anything but character descriptions (and maybe symbols?) for other Indian languages. Do you think it makes sense for us to just commit the files directly to NVDA Git or do you think they should be managed by srt even so?

Straight into git, otherwise its creating extra work for us on the translation system side without a real benefit. Esentially it immidiately becomes an unmaintained language/not in the translation system. If there is renewed intrest in doing complete translations then the files for those specific languages can be copied back into the svn.

I have not committed Hindi/punjabi character descriptions, as they need to be checked against those already being maintained. As Jamie says:

Because we already have Hindi translators on board, they need to be managing/approving any changes.

If you wish to contribute on more than character descriptions for those languages, you should join the team for that language, and use the translation system.

Also, because this is a full translation,

[I am not able to find the full translation, where should I be looking?

Replying to [comment:4 siddhartha_iitd](...]

Sorry):

Please check the branch in_t4410 at the below mentioned url:

https://manish_agrawal@bitbucket.org/manish_agrawal/nvda.git

the in_t4410 contains other changes, not purely the work related to this ticket, so I did not use it, just copied the files and made a new commit.

I am not sure who wrote this document: http://blog.vardhan.net/2013/10/git-flow-for-nvda-india-branch.html

but basically in_t4410 should not contain work for other tickets, and next/in_next should never be merged into it, you should be merging in_t4410 into in_next.

In any case, please test nvda next snapshot 11463 or later.

I will leave Hindi/Punjabi character descriptions comparison/merge to be sorted out by Him Prasad, and I expect to see any changes in the translation system, since this requires minimum effort for everyone.

nvaccessAuto commented 9 years ago

Comment 14 by MHameed on 2015-01-10 10:39 In [728045fa98120b2f53d30a14e6720f3ecd97d733]:

 revision="728045fa98120b2f53d30a14e6720f3ecd97d733"
 Merge branch 't4410' into next. Incubates #4410

Changes: Added labels: incubating

nvaccessAuto commented 9 years ago

Comment 15 by James Teh <jamie@... on 2015-01-30 05:25 In [6e42c0ccf51b5c3a60e7fa6fbeafc785ae64499f]:

Add character descriptions for some Indian languages: Assamese, Bengali, Gujarati, Konkani, Malayalam, Manipuri and Telugu.

Fixes #4410.

Changes: Removed labels: incubating State: closed

nvaccessAuto commented 9 years ago

Comment 16 by jteh on 2015-01-30 05:26 Changes: Milestone changed from None to 2015.1