---
id: jaconv
version: "0.5.0"
license: MIT License
license_treatment: permissive
maintenance: active
---

# jaconv — Pure-Python Japanese character interconverter for Hiragana, Katakana, Hankaku, Zenkaku and more

License: permissive · Maintenance: active · Downloads: 3.6M/mo

## What it is and what it does

jaconv is a pure-Python library for converting between Japanese character representations. It handles transformations between Hiragana and Katakana scripts, conversion between half-width (Hankaku) and full-width (Zenkaku) character forms, and bidirectional transliteration between Japanese kana and romanized alphabet. The library also provides Unicode normalization tailored for Japanese text, including handling of small kana characters and phonetic format conversion for speech synthesis tools like Julius.

The package is designed for text preprocessing and normalization workflows in Japanese natural language processing. It has no external dependencies and runs on Python 2.7 through 3.14, making it straightforward to integrate into existing pipelines. Each conversion function accepts optional parameters to control which character types are affected, allowing fine-grained control over transformations.

Use it for:

- Normalize Japanese text input by converting mixed character widths to consistent full-width or half-width forms before processing.
- Prepare Japanese text for search or indexing by converting between Hiragana and Katakana to handle user input variations.
- Convert user-entered romanized text (romaji) to Hiragana or Katakana for Japanese input method simulation.
- Generate phonetic representations for Japanese text in Julius speech synthesis format.
- Preprocess Japanese datasets for machine learning by standardizing character representation across training data.

## Worth the install?

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

jaconv converts between Japanese character types: Hiragana, Katakana, half-width (Hankaku) and full-width (Zenkaku) characters, plus transliteration to and from romanized alphabet.

Yes. jaconv is a stable, actively maintained library with no dependencies, permissive licensing, and a focused scope for Japanese character conversion. Install it if you work with Japanese text preprocessing, normalization, or transliteration. The recent release cycle and broad Python version support make it reliable for production use.

## Install

pip install jaconv
uv add jaconv
poetry add jaconv

## Installing jaconv

Before you install: Low friction: pure Python wheel with no runtime dependencies. Actively maintained as of 2026-02-08 with recent support for Python 3.13 and 3.14.

License in practice: MIT License permits commercial and private use with minimal restrictions; suitable for most projects.

Quickstart:

pip install jaconv

import jaconv

# Hiragana to Katakana
jaconv.hira2kata('ともえまみ')
# => 'トモエマミ'

# Half-width to full-width
jaconv.h2z('ﾃｨﾛ･ﾌｨﾅｰﾚ')
# => 'ティロ・フィナーレ'

# Hiragana to romanized alphabet
jaconv.kana2alphabet('じゃぱん')
# => 'japan'

## Package facts

- License: MIT License (permissive)
- Python support: unspecified
- Install friction: low
- Maintenance: active
- Downloads: 3.6M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

japanese character conversion, hiragana katakana converter, half-width full-width kana, japanese text normalization, romaji transliteration, japanese preprocessing, kana to alphabet, japanese-nlp, text-normalization, transliteration

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