Giới thiệu công cụ
Many systems still choke on accented characters: old databases, SMS gateways, file systems, usernames and URL slugs. This tool decomposes text with Unicode NFD normalization and removes the combining diacritical marks from Latin and Greek letters, plus Arabic harakat. Strict ASCII mode goes further and transliterates letters that are not built from marks, such as ß → ss, æ → ae, ø → o, ł → l and đ → d, and converts smart quotes, dashes and ellipses. Devanagari and Thai vowel signs are never stripped, because removing them would destroy the words.
Như mọi công cụ trên LanguageHelper, Bỏ dấu và ký tự đặc biệt chạy hoàn toàn trong trình duyệt của bạn. Văn bản không bao giờ được tải lên, lưu trữ hay dùng để huấn luyện, nên an toàn cho tài liệu mật, dữ liệu khách hàng và tin nhắn cá nhân. Không cần tài khoản hay cài đặt, và công cụ vẫn hoạt động ngoại tuyến sau khi đã tải.
Cách hoạt động
- Paste or type your text — the output updates instantly.
- Enable “Strict ASCII” for pure 7-bit output.
- Keep optional marks in other scripts (ё, ї, nukta, Thai tones) if you only want Latin accents removed.
- Copy or download the cleaned text.
Trường hợp sử dụng
- Create clean URL slugs, file names and usernames.
- Import names into legacy systems that only accept ASCII.
- Normalize text for search, sorting and deduplication.
- Send SMS messages without Unicode characters that increase cost.
Ví dụ
Trước
Crème brûlée à São Paulo
Sau
Creme brulee a Sao Paulo
Trước
Łódź, Straße, “quote”
Sau
Lodz, Strasse, "quote"
Mẹo để có kết quả tốt nhất
Use default mode when you only need to drop accents but keep meaningful letters such as Polish ł or German ß; use strict ASCII when the target system rejects anything outside 7-bit ASCII. For search indexes, normalize both the stored text and the query in the same way. Remember that removing accents can change meaning in some languages, so keep the original for display.
Câu hỏi thường gặp
What is the difference between default and strict mode?
Default mode only removes combining marks, so “ł” or “ß” stay as they are because they are distinct letters. Strict ASCII mode replaces every non-ASCII Latin character with its closest ASCII equivalent.
Does it work with Vietnamese?
Yes. Vietnamese letters with stacked marks, such as “ệ” or “ữ”, are decomposed and reduced to their base letters, and “đ” becomes “d” in strict mode.
Can I remove Arabic tashkeel?
Yes. Arabic short-vowel marks (fatha, kasra, damma, shadda, sukun, tanwin) are removed while the base letters stay intact.
Will it break Hindi or Thai text?
No. Vowel signs in Devanagari and Thai and Japanese kana are never touched. By default only optional marks are removed: Devanagari nukta (ज़ → ज) and chandrabindu, Thai tone marks, and Cyrillic ё/ї and stress accents (й stays). Tick “Keep optional marks in other scripts” to leave those too.
Is there a size limit?
No practical limit. Processing is instant and local, so even long documents are converted in milliseconds without uploading anything.
Is this the same as slugifying?
It is the first step. A URL slug also lower-cases text and replaces spaces and punctuation with hyphens; remove accents here with strict ASCII on, then apply your slug rules.