Spell Checker Mn

Spell Checker Mn detects misspellings in English text and returns curated correction suggestions. It targets writers, students, editors, and content teams who produce emails, reports, documentation, blogs, and web content. It does not perform grammar checks or style editing; instead it prioritizes spelling accuracy and token-level corrections while preserving meaning, punctuation, and formatting. The tool supports US and UK English conventions and can be extended with a custom user dictionary to accommodate domain-specific terms, acronyms, and brand names without forcing replacements.
Conceptually, the system uses a two-layer approach: a fast dictionary lookup flags non-dictionary tokens, then a contextual reranking stage using a lightweight n-gram model to prefer edits that fit neighboring words. It computes candidate corrections using Levenshtein distance and ranks them by score, frequency, and lexical validity. It preserves capitalization and punctuation, and can treat proper nouns specially when enabled, to minimize disruptive changes to names, titles, and technical terms. When no safe correction exists, it returns the original token with a low-confidence suggestion.
Core features include configurable language variants, a customizable dictionary, batch processing for multiple texts, and a clean output that pairs corrected_text with a structured suggestions list. Typical use cases include proofreading emails, draft articles, course materials, product descriptions, and user guides. The unique value lies in combining locale-aware spelling rules with non-intrusive edits, enabling rapid QA and consistent writing quality across teams. Outputs support downstream processing, exporting the corrected text or a detailed report of changes for review and audit.