---
id: pykakasi
version: "2.3.0"
license: GPL-3.0-or-later
license_treatment: copyleft
maintenance: dormant
---

# pykakasi — Kana kanji simple inversion library

License: copyleft · Maintenance: dormant · Downloads: 1.7M/mo

## What it is and what it does

Pykakasi is a Python NLP library that converts Japanese text into romanized form (rōmaji). It handles hiragana, katakana, and kanji characters, producing multiple output forms: kana (katakana), hiragana, and rōmaji (Latin alphabet). The library is based on algorithms from the kakasi C library and supports NFC-form characters.

Typical use is to pass Japanese text to a kakasi instance and iterate over the conversion results, which include the original character, its kana and hiragana forms, and its romanized representation. It depends on jaconv for character conversion, deprecated for API management, and importlib-resources for resource loading.

Use it for:

- Convert Japanese text to romaji for display in systems that require Latin characters
- Extract phonetic readings (hiragana/katakana) from kanji for language learning or accessibility
- Process Japanese input in NLP pipelines that need standardized romanized output
- Generate furigana-like output for mixed Japanese text (kana + kanji) in documents

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Transliterates Japanese text (hiragana, katakana, kanji) into rōmaji (Latin alphabet) and provides kana conversions.

Yes, if you need Japanese-to-romaji conversion and can tolerate dormant maintenance. The package is Production/Stable, has no known vulnerabilities, low install friction, and supports current Python versions (3.8–3.13). The 781-day gap since last release is notable but not disqualifying for a mature, narrowly-scoped library. GPL-3.0-or-later licensing is a blocker for proprietary closed-source projects.

## Install

pip install pykakasi
uv add pykakasi
poetry add pykakasi

## Installing pykakasi

Before you install: Low install friction; pure Python wheel. Maintenance is dormant (781 days since last release), though the package is marked Production/Stable and has no known vulnerabilities.

License in practice: GPL-3.0-or-later copyleft license. Derivative works and distributions must be released under GPL3+ or compatible terms; proprietary use requires separate licensing.

Quickstart:

import pykakasi
kks = pykakasi.kakasi()
text = "かな漢字"
result = kks.convert(text)
for item in result:
    print(item['orig'], item['hepburn'])

Verify before relying:

- Whether dormant maintenance status (781 days) affects reliability for current Python versions despite Production/Stable classifier
- Performance characteristics on large Japanese text volumes or real-time processing scenarios

## Package facts

- License: GPL-3.0-or-later (copyleft)
- Python support: supports_current
- Install friction: low
- Maintenance: dormant
- Downloads: 1.7M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

japanese text to romaji, hiragana katakana converter, kanji transliteration, japanese nlp library, kana romaji conversion, japanese character processing, japanese-nlp, text-processing

[View on SkillFed](https://skillfed.io/packages/pykakasi) · [View on PyPI](https://pypi.org/project/pykakasi/)