Text to Binary Converter

Type or paste text to see it as binary, one 8-bit byte per group, with a table showing how each character was encoded. Accented letters and emoji are encoded as UTF-8, so the result converts back exactly.

Any text works, including accented letters and emoji.

Binary (UTF-8)
Result appears as you type

How text is converted to binary

Computers store text as numbers. Each character has a code point (A is 65, a is 97, a space is 32), and an encoding decides how that number is stored as bytes. The web uses UTF-8 almost everywhere, so that is what this tool uses.

In UTF-8, the 128 ASCII characters (English letters, digits and common punctuation) take one byte each, so their binary is just the code written in 8 bits. Other characters take two to four bytes: é is 2 bytes, € is 3 and most emoji are 4.

Example 1: Hi → 01001000 01101001

Each character becomes one or more bytes

Every character has a code number. For ordinary English letters, digits and punctuation, UTF-8 stores it in a single byte, written as 8 bits.

CharacterCode pointUTF-8 bytes (hex)Binary
HU+00484801001000
iU+00696901101001

2 bytes

Example 2: Café → 01000011 01100001 01100110 11000011 10101001

Each character becomes one or more bytes

Every character has a Unicode code point. UTF-8 stores ASCII characters in 1 byte and other characters in 2 to 4 bytes. Each byte is then written as 8 bits.

CharacterCode pointUTF-8 bytes (hex)Binary
CU+00434301000011
aU+00616101100001
fU+00666601100110
éU+00E9C3 A911000011 10101001

5 bytes

Where this comes up

  • Computer science homework and exams that ask you to encode a word in ASCII or binary.
  • Checking how many bytes a string really takes, for example when a database column or SMS limit counts bytes rather than characters.
  • Puzzles, escape rooms and messages hidden in binary.

Frequently asked questions

Why does é turn into two bytes?

é is not an ASCII character. Its code point is 233, and UTF-8 stores any code point above 127 in two or more bytes, so é becomes 11000011 10101001. Older encodings such as Latin-1 used a single byte (11101001), which is why text sometimes looks garbled when the wrong encoding is used.

Is this ASCII or UTF-8?

Both, for plain English text. UTF-8 was designed so that the first 128 characters are stored exactly as in ASCII, one byte each. Only characters outside ASCII use more bytes.

What is the binary for a space?

A space is character 32, which is 00100000 in binary.

Is my text sent anywhere?

No. The conversion runs in your browser and nothing you type is uploaded.

Related tools

Local Processing
End-to-End Encryption
Works Offline

Building Something Beyond This Tool?

We help teams design and build reliable web and mobile applications backed by well-structured APIs.