Chinese Text & Pinyin Toolkit

Run common Chinese localization and text-normalization operations in one batch Actor. It is designed for content catalogs, search indexes, URL generation, CRM cleanup, AI pipelines, and multilingual publishing.

Features

• Pinyin with tone marks, tone numbers, no tones, and initials.
• Simplified Chinese to Traditional Chinese.
• Traditional Chinese to Simplified Chinese.
• Unicode NFKC, full-width to half-width, punctuation, and whitespace normalization.
• Chinese or mixed-language text to URL-safe pinyin slugs.
• Chinese numerals to Arabic numbers.
• Arabic numbers to lowercase Chinese numerals.
• Numeric amounts to uppercase RMB wording.
• Up to 1,000 records and 100,000 total input characters per run.

Input

Use texts for a simple batch:

{
  "texts": ["Chinese title", "Traditional Chinese text", "1234.56"],
  "operations": ["pinyin", "simplified", "traditional", "normalize", "slug"],
  "pinyinSeparator": " "
}

The Apify input form contains real Chinese examples. Use records instead when each text needs a caller-defined ID:

{
  "records": [
    {"id": "product-1", "text": "Chinese product title"},
    {"id": "invoice-1", "text": "1234.56"}
  ],
  "operations": ["slug", "rmbUppercase"]
}

When records is supplied, it takes precedence over texts.

Operations

• pinyin: returns toneMarks, toneNumbers, plain, and initials.
• simplified: converts Traditional Chinese to Simplified Chinese.
• traditional: converts Simplified Chinese to Traditional Chinese.
• normalize: normalizes width, punctuation, Unicode compatibility forms, and whitespace.
• slug: creates a lowercase ASCII pinyin slug.
• chineseToNumber: converts one Chinese numeral expression to a number.
• numberToChinese: converts one Arabic number to lowercase Chinese wording.
• rmbUppercase: converts one Arabic amount to uppercase RMB wording.

If a number operation receives nonnumeric text, the item status becomes partial, an operation-specific error is returned, and other selected operations still succeed.

Notes

Pinyin can be context-dependent for polyphonic characters. The Actor uses phrase-aware dictionaries but cannot infer every proper name or specialist pronunciation. Simplified and Traditional conversion uses OpenCC dictionaries and is linguistic conversion, not human translation.

This Actor uses the MIT-licensed pypinyin and cn2an libraries and the Apache-licensed OpenCC Python reimplementation.