Part of our project of digitizing Jewish liturgy is to provide a resource to convert the consonants and vowels of Hebrew into any other script. Ultimately this will be a standard feature in the web application we are building to help folk craft their own siddur, machzor, bentscher or other useful prayer book. Our lead developer, Efraim Feinstein, recently managed to put most of the pieces together to accomplish this, a milestone for the Open Siddur Project.
There is no single standard for Hebrew transliteration. In our demo you can transliterate Hebrew text in eight different ways originally set out in the following sources:
- Rules of Transcription from Hebrew Script to Latin Script (Academy of the Hebrew Language, 2007)
- International Phonetic Alphabet (2005, as used by Wikipedia)
- The SBL Handbook of Style (Society of Biblical Literature, 1999)
- Romanization Table for Hebrew and Yiddish (The American Library Association/Library of Congress, 1997)
- Concise Dictionary of the Words in the Hebrew Bible with their Renderings (James Strong, 1890)
- Coding for Transliteration of Hebrew (Michigan-Claremont, 1984)
- An approximation of Modern Israeli Hebrew pronunciation by Open Siddur lead developer, Efraim Feinstein (2010)
- An approximation of Modern Ashkenazi Hebrew pronunciation by Aharon Varady (2010)
Currently, the demonstration only provides romanization — the transliteration of Hebrew to a Latin script. By incorporating additional transliteration standards for additional scripts, we will be able to convert Hebrew to Greek, Cyrillic, Amharic, etc. (and vice versa). The tables are not fixed, and we can change them if bugs are found or better ways are suggested. Eventually, we will be implementing a table editor to allow editing the tables, creating, and of course, sharing new ones. For now, if you would like to add a transliteration standard to our database, take a look first at these examples.
The source code for this transliterator is open source, LGPL licensed, so you are free to take this and use it in your web application or website as well. Join us, and help make this a spectacular resource for everyone.
The form below provides a demonstration of this open source technology. Try it with some Hebrew now! If you don’t have any handy, try transliterating this phrase from the opening of the Amidah:
אֲדֹנָי שְׂפָתַי תִּפְתָּח וּפִי יַגִּיד תְּהִלָּתֶךָ׃
or this verse from Tzephaniah, with all the letters of the aleph-bet:
לָכֵן חַכּוּ לִי נְאֻם יְהוָה לְיוֹם קוּמִי לְעַד כִּי מִשְׁפָּטִי לֶאֱסֹף גּוֹיִם לְקָבְצִי מַמְלָכוֹת לִשְׁפֹּךְ עֲלֵיהֶם זַעְמִי כֹּל חֲרוֹן אַפִּי כִּי בְּאֵשׁ קִנְאָתִי תֵּאָכֵל כָּל הָאָרֶץ׃
“How to Transliterate Hebrew Text with the Latin Alphabet” is shared by The Hierophant with a Creative Commons Attribution-ShareAlike 3.0 Unported license.
great tool. I really like that there is the siddur-type-user-friendly transliteration and also the I-am-presenting-to-my-linguist-friends version (several, really). Thanks!
Thought you (and your linguist friends) might like an International Phonetic Alphabet transliterator, so I added one :)
Thanks for the IPA — it’s important to remember though that the IPA encodes a particular pronunciation, depending on which characters you pick; in the abstract it’s an objective system, but you have to make the decision whether to transliterate ח with /ħ/ or /x/, for example, or whether to transliterate ḥolam as /o/, /ou/, /o:/, /oj/, /ej/, etc.
Thanks Steg. I used the IPA table used by Wikipedia — and I should’ve noted that.
Efraim is going to be adding a feature for folks to modify the tables and create new ones on the fly. Ultimately, folk will be able to share their tables within the Open Siddur application as with all other content/layout template/etc..
This is lovely! Great work, I am seriously impressed with this code’s functionality.
Transliterating is wonderfully useful for catching shin dots in place of holam, and kamatz gadol/katan when proofreading transcriptions!
Two issues:
-I get unexpected transliterations (missing a’s) of the Name, written in yod-yod form, either unvocalized or with schwa-kamatz (U+05B8)
-pasting text with X middle-click in “Enter the text to transliterate here:” alone is not trapped by the script as a content update. A futher editing action (even pressing delete after the last char) is needed, else either the previous transliteration is presented again, or the empty content error pops up.
Anything new on the user edited tables front?
Amir, make sure to see Efraim’s response below.
There is a (completed) new format for user-edited tables. There is no “editor” as-such, but the XML format is rather simple for hand-editing.
It’s also better to ask questions on the opensiddur-tech email list than here. They’re more likely to be seen and answered there.
[...] tested and then made public our transliteration engine with eight transliteration tables to choose from including International [...]
[...] lead developer, Efraim Feinstein and all the developers who had a hand in contributing to our transliteration engine, the source of the transliterations used in the four verses of the Megillah that are communally [...]