Hebrew Text Tools#

Pure Python utilities for Hebrew text processing. No dependencies, works offline.

Quick Start#

bash

# Transliterate Hebrew to Latin
hebrew-tools "שלום עולם"
# Output: shalom olam

# Calculate gematria
hebrew-tools -g "בראשית"
# Output: 913

# Remove nikud (vowel points)
echo "שָׁלוֹם" | hebrew-tools -n
# Output: שלום

# List letter names
hebrew-tools -l "אבג"
# Output: Alef, Bet, Gimel

# Format number as Hebrew letters
hebrew-tools -N 613
# Output: תרי"ג

Commands#

CommandDescriptionExamplehebrew-tools <text>All transformationshebrew-tools "שלום"``hebrew-tools -t <text>Transliterate onlyhebrew-tools -t "תורה" → "torah"hebrew-tools -g <text>Gematria onlyhebrew-tools -g "חי" → 18hebrew-tools -n <text>Remove nikudhebrew-tools -n "בְּרֵאשִׁית" → "בראשית"hebrew-tools -l <text>Letter nameshebrew-tools -l "אב" → [Alef, Bet]hebrew-tools -r <text>Reverse RTLhebrew-tools -r "שלום" → "מולש"hebrew-tools -N <num>Number to lettershebrew-tools -N 26 → "כו"

Output Format (Default)#

text

original: שלום
has_hebrew: True
transliteration: shalom
no_nikud: שלום
gematria: 376
letter_names: [Shin, Lamed, Vav, Mem]

Features#

Transliteration#

• Ashkenazi-style pronunciation
• Handles all Hebrew letters including sofit (final forms)
• Shin/Sin distinction (dot right/left)
• Dagesh handling

Gematria#

• Standard Mispar Hechrachi values
• Supports all Hebrew letters including sofit forms
• Works with or without nikud

Nikud Removal#

• Removes all Hebrew vowel points and cantillation marks
• Preserves base letters
• Handles composite characters

Letter Names#

• Returns English names for each Hebrew letter
• Sofit forms identified (e.g., "Mem Sofit")
• Non-Hebrew characters preserved

Hebrew Number Formatting#

• Converts integers to Hebrew letters (Gematria style)
• Standard abbreviations (e.g., תרי"ג for 613)
• Range: 1–999

Python API#

python

from scripts.hebrew_tools import transliterate, gematria, remove_nikud

# Transliterate
print(transliterate("תורה"))  # "torah"

# Gematria
print(gematria("חי"))  # 18

# Remove nikud
clean = remove_nikud("בְּרֵאשִׁית")
print(clean)  # "בראשית"

Limitations#

• Transliteration is Ashkenazi-style; Sephardi variants not yet supported
• Hebrew number formatting only supports 1–999
• RTL reversal is basic (word-level, not character-level for mixed text)