Binary Code Translator

A bit is the smallest unit. Eight bits make a byte, and a byte is the usual packet for one ASCII character or one step of a UTF-8 sequence. The pattern 01000001 is not a secret code. It is the number 65 written in base 2, and 65 is the ASCII slot for the letter A. Flip the lowest bit and you get 01000000, which is 64, the character @ in ASCII. That is the whole trick: bits are place values, not Morse-style symbols.

People say “binary code” for a few different things. Machine instructions are binary. Color channels are binary. Network packets are binary. This page is a binary code translator for text: it writes characters as 0s and 1s, and it reads those bits back as characters. It does not disassemble programs or decode raw image files. If the input is meant to be a sentence, pick an encoding and convert. If the input is a firmware dump, you need a hex editor, not this tool.

Why bother looking at bits at all? Because every file on disk is bits, and a surprising number of homework problems, CTF puzzles, and “secret” social posts are just ASCII or UTF-8 written out as 0 and 1. Once you can read a byte, those puzzles stop being puzzles. The translator on this page exists so you can check an answer in a few seconds instead of adding powers of two by hand.

Read the byte from the left. The leftmost bit is worth 128, then 64, 32, 16, 8, 4, 2, and 1. Add the values where the bit is 1. For 01000001 that is 64 + 1 = 65. Character 65 in ASCII and UTF-8 is A. The same walk works for every other byte: 01100001 is 97, which is a; 00110000 is 48, which is the digit 0. You do not have to memorize a chart if you can add eight numbers.

A short example: the word hi in ASCII or UTF-8 is 01101000 01101001. The first byte is h (104), the second is i (105). Add a space and you get a third byte, 00100000. Newlines are bytes too — a Unix line break is 00001010 (LF). If you paste a paragraph and the binary looks longer than you expected, count the spaces and line breaks; they are real characters.

UTF-8 does not change the English letters. A, a, and 0 still occupy one byte, with the same bit patterns as ASCII. The extra bytes appear only when a character sits outside that range. é becomes two bytes (11000011 10101001). A typical emoji becomes four. That is why the same sentence can produce a longer bit string after you add one accented name or a smiley.

Spaces are optional. The translator strips whitespace, then walks the string in 8-bit steps. 01000001 and 01000001 with gaps both decode to A. What it cannot guess is encoding. If someone encoded café as UTF-8 and you decode as ASCII, the fourth character is outside ASCII. If they used UTF-16 BE and you stay on UTF-8, you get noise or an error. Match the source, then convert.

A leftover nibble is the usual foot-gun. Seven bits at the end of a paste means a byte was cut off in a screenshot or a chat wrap. Nine bits means an extra 0 or 1 snuck in. The inspector under the output flags an incomplete group so you can fix the paste instead of staring at a blank result. Also watch for look-alike characters: a capital O, a lowercase L, or a full-width １ from a PDF will not count as bits.

When the bits came from a programming exercise, they are almost always ASCII or UTF-8 in 8-bit groups. If every other byte is 00000000, switch encodings before you assume the tool is broken.

The first 32 slots are control codes. You will meet a few of them: 9 is tab, 10 is line feed, 13 is carriage return. Printable characters start at 32 (space) and run through 126 (tilde). Digits 0–9 are 48–57. Uppercase A–Z are 65–90. Lowercase a–z are 97–122. The 32-step gap between A and a is why flipping one bit (01000001 vs 01100001) changes case. That is not a coincidence; the table was built that way.

ASCII stops at 127. It has no é, no 你, and no emoji. Early systems stuffed extra letters into the unused high bit and called the result Latin-1, Windows-1252, or a dozen other 8-bit pages. Those pages disagree with each other, which is why a file that looks fine on one computer turns into mojibake on another. Unicode replaced that mess. UTF-8 keeps the ASCII bytes identical and uses extra bytes only when needed, so English text is both “ASCII” and “UTF-8” at the same time.

If a course asks you to “convert to binary using ASCII,” they almost always want one 8-bit group per character, high bit 0, letters as in the table above. They rarely want UTF-16. When in doubt, encode the sample word hello and check against 01101000 01100101 01101100 01101100 01101111. If that matches, you are on the encoding the assignment expects.

Use ASCII for basic English text, UTF-8 for web text and emoji, or UTF-16 BE when the source specifies big-endian UTF-16. The Unicode Standard, Chapter 2 defines these encoding forms and their byte-order conventions.

UTF-16 BE always writes at least two bytes. English A is 00000000 01000001 — the same 65, padded on the left. Characters above U+FFFF use a surrogate pair (four bytes). This translator does not emit a byte-order mark. If you need a BOM, add 11111110 11111111 (UTF-16 BE) yourself; most web and homework cases should not include one.

Two things this translator is not. It is not a calculator for two’s-complement arithmetic — it will not tell you that 11111111 is −1 as a signed byte. And it is not a packet sniffer. Ethernet frames and JPEG files contain binary that is not text. If you paste those bits, you may get a few printable characters and a lot of errors. Use a hex dump for files; use this page for messages that were meant to be read as text.

Source: Unicode Standard, Chapter 2 · Updated October 3, 2026