Count word/ngram frequency and density for SEO copy review — Chinese 2-4 char n-grams + English tokens with stopword filtering.
Paste a chunk of text. The tool detects whether the content is mostly Chinese or English and applies the right tokenizer: Chinese is sliced into overlapping 2-, 3- and 4-character n-grams; English is split on non-letter boundaries, optionally lowercased/stemmed. Common stopwords (的/了/是, the/a/is) are filtered out by default. The Top 20 table shows each keyword's count and density (count / total tokens). Density is a rough SEO signal — combine with semantic judgment.