hebrew-transliteration

A JavaScript package for transliterating Hebrew

install

npm

npm install hebrew-transliteration

local

You will need to have node installed.

Download or clone this repository

cd hebrew-transliteration
npm install
npm run build

example

const heb = require("hebrew-transliteration");
const transliterate = heb.transliterate;
transliterate("אֱלֹהִים");
// ʾĕlōhîm

DOCS

About

This is a JavaScript package for transliterating Hebrew.

It exports 3 functions:

1. transliterate() — the main function which transliterates Hebrew
2. remove() — removes taamim and optionally removes certain niqqudim
3. sequence() — sequences Hebrew characters according to the SBL Hebrew Font Manual

And it exports 2 classes:

1. Text — the Text class from the havarotjs package
2. Schema — a schema for transliterating Hebrew

Functions

transliterate()

Takes a string or Text, and optionally a Schema or Partial<Schema>

heb.transliterate("אֱלֹהִים");
// ʾĕlōhîm

If no Schema is passed, then the package defaults to SBL's academic style.

You can pass in a Partial<Schema> that will modify SBL's academic style:

heb.transliterate("שָׁלוֹם", { SHIN: "sh" });
// shālôm

There are premade schemas as well.

const brillAcademic = require("hebrew-transliteration/schemas").brillAcademic;

heb.transliterate("בְּבֵית", brillAcademic)
// bᵉḇêṯ

Note: schemas are not endorsed by publishers.

If you need a fully customized transliteration, it is best to use the Schema constructor:

const schema = new heb.Schema({
  ALEF: "'",
  BET: "B",
  ...
  QAMETS: "A",
  ...
}) // truncated for brevity

heb.transliterate("אָ֣ב", schema)
// 'AB

remove()

Takes string and RemoveOptions.

The default removes accents (i.e. characters called HEBREW ACCENT) and metheg and rafe.

heb.remove("שָׂרַ֣י אִשְׁתְּךָ֔");
// שָׂרַי אִשְׁתְּךָ;

The RemoveOptions can be customized.

heb.remove("שָׂרַ֣י אִשְׁתְּךָ֔", { SHIN_DOT: true, SIN_DOT: true });
// שָרַ֣י אִשְתְּךָ֔

Note: unlike a Schema where a Partial<Schema> extends the default, RemoveOptions does not accept a Partial<RemoveOptions>.

All properties for RemoveOptions can be found in the source.

There are some preset options availables as well.

const opts = require("hebrew-transliteration/removeOptions");
heb.remove("שָׂרַ֣י אִשְׁתְּךָ֔", opts.all);
// שרי אשתך, וימצאו

sequence()

Takes a string. Returns a string of properly sequenced characters according to the SBL Hebrew Font manual following the pattern of: consonant - dagesh - vowel - ta'am

heb.sequence("\u{5D1}\u{5B0}\u{5BC}\u{5E8}\u{5B5}\u{5D0}\u{5E9}\u{5B4}\u{5C1}\u{596}\u{5D9}\u{5EA}");
//           "\u{5D1}\u{5BC}\u{5B0}\u{5E8}\u{5B5}\u{5D0}\u{5E9}\u{5C1}\u{5B4}\u{596}\u{5D9}\u{5EA}"

Classes

Text

The Text class from the havarotjs package.

This class is used by transliterate() internally to syllabify Hebrew text, but it is exposed as well.

const text = new heb.Text("הֲבָרֹות");
text.syllables;
// [
//    Syllable { original: "הֲ" },
//    Syllable { original: "בָ" },
//    Syllable { original: "רֹות" }
//  ]

If a Text is passed into transliterate() instead of a string, then the syllabification from the Text class is used. If a string is passed in, then syllabification come from the options passed into the Schema. See more under syllabification.

Schema

A Schema is used to define a schema for transliteration. See the Schema source for all available properties.

The Schema can be divided into a few categories.

1) Syllabification

The options used for syllabifying Hebrew text can be found here

Differences between Text and Schema

There are 5 options for syllabification that are the same as the ones used by the Text class.

Read more about the syllabification options for the Text and a higher level overview

Precedence of Text over Schema

The syllabification options set by Schema are used if a string is passed into transliterate(). If a Text is passed into transliterate() instead of a string, then the syllabification from the Text class is used:

// using default
heb.transliterate("חָכְמָ֣ה"); // ḥokmâ

// using Schema for syllabification
heb.transliterate("חָכְמָ֣ה", { qametsQatan: false }); // ḥākǝmâ

// using Text for syllabification
heb.transliterate(new heb.Text("חָכְמָ֣ה", { qametsQatan: false })); // ḥākǝmâ

// using Schema and Text — Text takes precedence
heb.transliterate(new heb.Text("חָכְמָ֣ה", { qametsQatan: true }), { qametsQatan: false }); // ḥokmâ

Note: qametsQatan only converts a regular qamets character; if a qamets qatan character is used, it will always be a qamets qatan.

2) Characters

Most Schema properties are for defining single Hebrew characters:

heb.transliterate("אָ", { ALEF: "@", QAMETS: "A" });
// @A

3) Orthographic Features

Some properties are for defining common Hebrew orthographies for:

BeGaDKePhaT

There are properties for the digraphs of BeGaDKePhaT letters:

• BET_DAGESH
• GIMEL_DAGESH
• DALET_DAGESH
• KAF_DAGESH
• PE_DAGESH
• TAV_DAGESH

Each one is the consonant character followed by the dagesh character (U+05BC).

These are helpful for distinguishing between spirantized forms.

heb.transliterate("בְּבֵית", { BET: "b" });
// bǝbêt

heb.transliterate("בְּבֵית", { BET: "v", BET_DAGESH: "b" });
// bǝvêt

Matres Lectionis

The following properties are for matres lectionis:

• HIRIQ_YOD
• TSERE_YOD
• SEGOL_YOD
• SHUREQ
• HOLAM_VAV
• QAMATS_HE
• SEGOL_HE
• TSERE_HE

heb.transliterate("פֶּה", { SEGOL_HE: "é" });
// pé

Others

There are other orthographic features:

• MS_SUFX — HEBREW LETTER QAMATS (U+05B8) and YOD (U+05D9) and VAV (U+05D5) יו◌ָ
• DIVINE_NAME — the full form of the divine name - יהוה
• DIVINE_NAME_ELOHIM — optionally, the form of the divine name pointed as ʾelōhîm (e.g. יֱהֹוִה)
• SYLLABLE_SEPARATOR — a syllable separator, usually an empty string
• DAGESH_CHAZAQ — if true, repeats the consonant with the dagesh, or can take a string

heb.transliterate("שַׁבָּת", { DAGESH_CHAZAQ: true });
// šabbāt

heb.transliterate("שַׁבָּת", { DAGESH_CHAZAQ: false });
// šabāt

heb.transliterate("שַׁבָּת", { DAGESH_CHAZAQ: "\u0301" });
// šab́āt

heb.transliterate("הָאָֽרֶץ", { SYLLABLE_SEPARATOR: "-" });
// hā-ʾā-reṣ

4) Others

Additional Features

The ADDITIONAL_FEATURES property is for defining non-typical Hebrew orthography, example:

heb.transliterate("הַזֹּאת", {
  ADDITIONAL_FEATURES: [
    {
      FEATURE: "cluster",
      HEBREW: "זּ",
      TRANSLITERATION: "tz"
    }
  ]
});
// hatzōʾt

• The orthography זּ is most often a doubling of the ZAYIN (i.e. 'z' with no dagesh, and 'zz' with a dagesh chazaq)
• In the Romaniote reading tradition, however, the ZAYIN is usually transliterated with 'z' (really 'ζ'),
• but a ZAYIN followed by a dagesh is transliterated as 'tz' (really 'τζ')

Each additional feature consists of 4 properties:

1. FEATURE — has three options:

• "cluster" — a cluster is any combination of a single character and optionally a dagesh and vowel.
• "syllable" — a syllable is any combination of a multiple characters and a single vowel and optionally a dagesh
• "word" — covers everything else

2. HEBREW — the Hebrew text to be transliterated
3. TRANSLITERATION — the text used to transliterate the Hebrew text, or a callback function
4. PASS_THROUGH — if true passes the characters of the result of the TRANSLITERATION callback to the be mapped to the schema. If TRANSLITERATION is a string, this does nothing. Default true.

Using a callback

A callback can be used TRANSLITERATION instead of just a string.

const heb = require("hebrew-transliteration");
const rules = require("hebrew-transliteration/dist/rules");

// use a callback to transliterate a vocal sheva with the same character as the next syllable
heb.transliterate("בְּרֵאשִׁ֖ית וַיַּבְדֵּל", {
  ADDITIONAL_FEATURES: [
    {
      // matches any sheva in a syllable that is NOT preceded by a vowel character
      HEBREW: "(?<![\u{05B1}-\u{05BB}\u{05C7}].*)\u{05B0}",
      FEATURE: "syllable",
      TRANSLITERATION: function (syllable, _hebrew, schema) {
        const next = syllable.next;
        // ensure type safety
        const nextVowel = next.vowelName === "SHEVA" ? "VOCAL_SHEVA" : next.vowelName;

        if (next && nextVowel) {
          const vowel = schema[nextVowel] || "";
          return syllable.text.replace(new RegExp("\u{05B0}", "u"), vowel);
        }

        return syllable.text;
      }
    }
  ]
});

// bērēʾšît wayyabdēl

:warning: this is an experimental ADDTIONAL_FEATURES; results may not always meet expectations.

Stress Marker

The STRESS_MARKER property is an optional property to indicate stress in transliteration.

It's properties are:

• location
• mark
• exclude (optional)

Example

heb.transliterate("מֶ֣לֶךְ", { STRESS_MARKER: { location: "after-vowel", mark: "\u0301" } });
// mélek

mark

The string used to mark stress (e.g. a combining acute accent (U+0301) )

location

The location has four options:

• "before-syllable"
• "after-syllable"
• "before-vowel"
• "after-vowel"

A combining mark (e.g. "\u0301") placed "after-vowel" will print on top of the vowel, and placed after a digraph will print on the second vowel.

heb.transliterate("בֵּ֣ית", {
  TSERE_YOD: "ei",
  STRESS_MARKER: { location: "after-vowel", mark: "\u0301" }
});
// beít

exclude

An optional property determining whether to exclude the mark on certain syllables.

It has three options

• undefined/"never"
• "single"
• "final"

Examples:

// undefined and "never" are the same
heb.transliterate("בֹּ֖קֶר י֥וֹם אֶחָֽד׃ ", {
  STRESS_MARKER: {
    location: "after-vowel",
    mark: "\u0301",
  }
});

// bṓqer yốm ʾeḥā́d

// exclude only single syllable words
heb.transliterate("בֹּ֖קֶר י֥וֹם אֶחָֽד׃ ", {
  STRESS_MARKER: {
    location: "after-vowel",
    mark: "\u0301",
    exclude: "single"
  }
});

// bṓqer yôm ʾeḥā́d

// exclude when accent is on the final syllable
// implicitly excludes single syllable words
heb.transliterate("בֹּ֖קֶר י֥וֹם אֶחָֽד׃ ", {
  STRESS_MARKER: {
    location: "after-vowel",
    mark: "\u0301",
    exclude: "single"
  }
});

// bṓqer yôm ʾeḥād

Live

Use it live at https://hebrewtransliteration.app

Contributing

See contributing