Yomi Text Analysis — Free Japanese Tokenizer API (No Key)

TL;DR

Yomi is a Japanese language morphological analysis web API. Submit Japanese text and get back tokenized results with part-of-speech tags, readings (yomi), base forms, and character types. Designed for Japanese language learning tools, text analysis, and NLP pipelines. Uses MeCab-compatible dictionaries and returns structured JSON. No API key required.

Quick start: https://yomi.text-analysis.com/api/tokenize?text=今日はいい天気です

No API key needed — free Japanese text analysis!

How to Use This API

1. Tokenize Japanese Text

Split Japanese text into individual tokens with linguistic data:

https://yomi.text-analysis.com/api/tokenize?text=今日はいい天気です

2. Get Detailed Analysis

Include readings, pronunciation, and conjugation info:

https://yomi.text-analysis.com/api/tokenize?text=食べ物を買いました&detailed=true

3. JavaScript — Parse Japanese Text

fetch('https://yomi.text-analysis.com/api/tokenize?text=日本語を勉強しています')
  .then(r => r.json())
  .then(data => {
    data.tokens.forEach(t => {
      console.log(`${t.surface} [${t.reading}] — ${t.pos}`);
      console.log(`  Base: ${t.base_form}, Type: ${t.char_type}`);
    });
  });

4. Python — Extract Nouns

import requests

text = '東京タワーで友達と写真を撮りました'
resp = requests.get(
    'https://yomi.text-analysis.com/api/tokenize',
    params={'text': text, 'detailed': 'true'}
).json()

nouns = [t for t in resp['tokens'] if t['pos'] == '名詞']
print('Nouns found:')
for n in nouns:
    print(f"  {n['surface']} ({n['reading']})")

Tokenize Japanese: https://yomi.text-analysis.com/api/tokenize?text=今日はいい天気です

Try This URL Now

Frequently Asked Questions

What linguistic data does each token include?: Each token includes surface (the text as written), reading (yomi/pronunciation), base_form (dictionary form), pos (part of speech), pos_detail (sub-classification), conjugation, and char_type (kanji, hiragana, katakana, etc.).
Does it handle proper nouns and place names?: Yes — the tokenizer recognizes common proper nouns, place names (東京, 大阪), and personal names. Accuracy depends on the dictionary entries available in the underlying MeCab dictionary.
Can it handle furigana output?: While the API doesn't directly output furigana-formatted HTML, the reading field provides the pronunciation for each token, which can be used to generate furigana annotations in your application.
What character encoding is required?: Send text as UTF-8. The API handles Japanese characters (kanji, hiragana, katakana), as well as mixed-language text with Latin characters and numbers.
Are there any text length limits?: The API supports reasonable text lengths for analysis. For very long documents, consider splitting the text into paragraphs or sentences and making multiple requests.
What dictionary does Yomi use?: Yomi is built on MeCab, a popular Japanese morphological analyzer. It uses the IPADIC dictionary for word segmentation and linguistic features, providing high accuracy for modern Japanese.

API Details

API URL: https://yomi.text-analysis.com/api/tokenize
Documentation: github.com/ookii-tsuki/yomi
Category: Text Analysis
Authentication: Not Required
Input Encoding: UTF-8

What You Can Build

Japanese language learning app showing word breakdowns and readings
Furigana generator for Japanese web content accessibility
Text analysis dashboard for Japanese writing statistics
Vocabulary extractor to build custom study lists from any text
Japanese search engine with morphological indexing capability