New Word Counting Algorithm for Non-Spaced Languages

Smartcat has a new word counting algorithm for languages that don't separate words with spaces: Chinese, Cantonese, Japanese, Thai, Burmese, Tibetan, Khmer, Lao, and Vietnamese. Instead of counting space-separated words, Smartcat estimates the word count from the number of characters, base letters, or syllables in each segment, using a ratio specific to each language or writing system. Word counts feed project statistics, quotes, and Smartwords charges, so the method matters for planning and billing.

Each segment is counted on its own and the totals are added together. A segment's main script is counted in its own unit, multiplied by the language's ratio, and rounded to the nearest whole word. Japanese weights hiragana and katakana at 0.30 and kanji at 0.60 because it mixes scripts. Words outside the main script, such as Latin words, numbers, and product names, count as one word each, every URL or email link counts as one word, and a segment with any text in its main script never counts as zero. For projects with a Chinese, Cantonese, or Japanese source, project statistics and the Excel export show alphabetic words, Asian characters, and a combined total in separate columns.

To learn more in our word counting help center article.