Leetspeak Translator

How it is calculated

In encode mode, each character is looked up in a fixed table: a→4, b→8, e→3, g→6, i→1, l→1, o→0, s→5, t→7, z→2. Capitals use the same entry as lowercase, so E and e both become 3. Every other character, including letters such as c, m or r, digits, spaces and punctuation, is copied unchanged and keeps its original case. In decode mode the table is reversed: 4→a, 8→b, 3→e, 6→g, 1→i, 0→o, 5→s, 7→t, 2→z, and everything else is left alone. Decoding is only a best guess. Both i and l become 1, so 1 always decodes to i, and the original capitalisation is lost. Empty text returns nothing, and any mode other than decode is treated as encode.

Common mistakes

• Expecting a round trip to give back the original text. Decoding turns every 1 into i, so Elite comes back as eiite, and capitals come back as lowercase.
• Expecting every letter to change. Only ten letters are in the table; c, m, n, r and others stay as they are, so the text is still partly readable.
• Decoding text that already contains genuine digits, such as a year or phone number. Those digits are converted to letters as well.
• Leaving the mode on decode while pasting normal text. Letters are then barely changed and the result looks like the input.
• Using the output as a password and treating it as secure. The substitutions are well known and add very little strength.