Braille translator

03 / Convention

How this variant works

1. Letters use standard 6-dot alphabet cells. Lowercase letters use one cell. Each uppercase letter is preceded by ⠠, the capital indicator.
2. A numeric indicator starts a digit run. The cell ⠼ changes a-j cells to digits 1-0. A comma or full stop can continue the run. Other supported punctuation and whitespace end it.
3. Grade 1 resolves a real ambiguity. A lowercase a-j after a digit, comma, or full stop gets ⠰, so 3b becomes ⠼⠉⠰⠃ and decodes exactly.
4. Ordinary spaces stay ordinary spaces. The no-dot pattern U+2800 looks blank but is not a Unicode space, so this tool rejects it instead of changing it silently.

Limits and safe use

• This is a small, deterministic UEB-aligned core, not a complete UEB translator. It does not implement contractions, capital-word or passage modes, accented letters, non-English codes, mathematics, music, BRF formatting, or embossing layout.
• Braille cells are context-dependent. In full UEB, ⠦ can be a question mark or an opening quotation mark. This defined subset always decodes it as a question mark and rejects unsupported indicator sequences.
• Only the listed print characters and their generated cells are accepted. Dot-7 or dot-8 patterns, the no-dot pattern U+2800, emoji, markup characters, and other scripts produce validation errors.
• Unicode glyphs are visual characters, not tactile output. Font support varies, and a copied result still needs suitable translation review and embossing equipment for formal accessible production.
• Input is limited to 2,000 Unicode code points and 4,000 UTF-16 code units. Browser text fields normalize CR and CRLF line endings to LF.
• Braille is a writing system, not encryption. It provides no secrecy and must not be used to hide sensitive information.