Leetspeak Converter

The leetspeak alphabet

Leet replaces letters with digits, punctuation and small clusters of symbols that look like the letter they stand in for. It is a homoglyph game, not a cipher: the reader is meant to recognise the shape, so the substitutions that stuck are the ones that survive being read quickly in a monospaced font. Only a handful are truly fixed. The rest vary by community, and several letters have no form anyone agrees on.

Reading a few of those out loud, since they are the ones people look up individually: T in leetspeak is 7, or + when you want the crossbar to read more clearly. P is |* or |o, and in heavier styles 9, which is also one of the two common forms for G. A is 4 or @. E is 3. S is 5 or $. B is 8. O is 0. Digits in the source text usually pass through untouched, because substituting a digit for a digit says nothing.

Basic leet and advanced leet

Basic leet swaps single characters and stays readable at a glance. l33t sp34k is still obviously "leet speak", which is the point: it marks the text without costing the reader anything. Most generators, this one included, default to that level.

Advanced leet stops being a character swap and starts being a dialect. Letters become multi-character constructions, so HACKER turns into |-|4X0R rather than H4CK3R, and the spelling itself changes as well as the glyphs.

• Multi-glyph letters: |-| for H, |\/| for M, \/\/ for W, |< for K, >< for X.
• The -xor and -0r agent suffix: hacker becomes haxor, then hax0r, then h4x0r.
• Plurals in z: warez, skillz, joolz. The s is not substituted so much as replaced.
• Set spellings that are not substitutions at all: teh for the, pwn for own, n00b for newbie, phear for fear.
• The -age and -ness nominalisers: pwnage, lootage, roxxorness.
• 1337 for leet itself, and 31337 for elite, which is where the number comes from.

Where leetspeak came from

Leet is short for elite. The dialect is usually traced to bulletin board systems in the 1980s, where boards that advertised restricted or elite access developed the substitutions among their users. Accounts differ on which board, which year and who first wrote what, and most of the early record is second hand, so any precise origin story is worth reading sceptically.

What the substitutions were for is better attested. They obscured words from crude keyword matching in board topics, file names and later chat filters. They signalled membership, since writing it fluently took practice. And once the first two uses faded they stayed on as style, moving through Usenet, IRC, warez scenes and eventually online games, where most people meet it today.

Decoding leetspeak back to English

Encoding is deterministic. Decoding is not, because the mapping is many-to-one and reversing it means guessing. Any tool that claims to translate leet back to English perfectly is overselling what the dialect allows.

• 1 is I or L, and sometimes just the digit one.
• 0 is O, or zero.
• 9 is G, or P in heavier styles, or a digit.
• 7 is T, or L, or a digit.
• 2 is Z, or the words to and too.
• 4 is A, or for.
• | on its own is I, L or a column of something else entirely.

A decoder gets useful results by trying the common substitutions and checking the candidates against a word list, which is roughly what a human reader does. Short strings, names and passwords are where it fails: l0l1 has several equally valid readings and no context to choose between them.

What this converter does

Light intensity substitutes vowels only, which keeps the text readable: A becomes 4, E becomes 3, I becomes 1 and O becomes 0. Heavy intensity applies the full map this tool carries, which is A to 4, B to 8, E to 3, G to 6, I to 1, L to 1, O to 0, S to 5, T to 7 and Z to 2. Letters outside that set, including U and the multi-glyph forms in the third column above, are left alone, so the output stays a single-character substitution rather than drifting into the advanced style.

Because I and L both map to 1, output is deliberately not round-trippable. That is a property of the dialect, not a limitation of the tool.

Related text tools

• Glitch text generator for stacked Unicode combining marks, a different way to mark up text.
• Upside down text for the same trick using rotated Unicode homoglyphs.
• Unicode converter to see what the substituted characters actually are.
• Morse code translator and NATO phonetic alphabet for substitution alphabets with real standards behind them.
• ROT13 and ROT-N cipher and Caesar cipher for actual letter-shifting ciphers.
• Case converter for the plainer transformations.