Morse code writes letters and digits as patterns of short signals (dots, spoken "dit") and long signals (dashes, spoken "dah"). This translator converts the English alphabet, digits and the most common punctuation marks to International Morse code, and reads code back into text. It is handy for memorizing the chart, understanding how a distress signal is built, preparing puzzles and escape-room clues, or warming up before amateur radio practice.
Input rules
- Text to Morse: letters are separated by one space and words by " / ". Line breaks and runs of spaces count as a single word break. Upper and lower case are treated the same.
- Morse to text: write dots as
.and dashes as-, separate every letter with a space, and put/between words. The middle dot and en or em dash characters are converted to dots and dashes automatically. - Anything that cannot be decoded is shown as "?" and the message below the box names the offending groups. The semicolon code
-.-.-.is a conventional one that does not appear in the ITU-R M.1677-1 list; the other punctuation marks (period, comma, colon, question mark, apostrophe, hyphen, parentheses, quotation mark, the double hyphen =, plus and @) match that recommendation.
The full International Morse chart (generated from this tool's data)
Letters
| Character | Code | Character | Code |
|---|---|---|---|
| A | .- | B | -... |
| C | -.-. | D | -.. |
| E | . | F | ..-. |
| G | --. | H | .... |
| I | .. | J | .--- |
| K | -.- | L | .-.. |
| M | -- | N | -. |
| O | --- | P | .--. |
| Q | --.- | R | .-. |
| S | ... | T | - |
| U | ..- | V | ...- |
| W | .-- | X | -..- |
| Y | -.-- | Z | --.. |
Digits
| Character | Code | Character | Code |
|---|---|---|---|
| 0 | ----- | 1 | .---- |
| 2 | ..--- | 3 | ...-- |
| 4 | ....- | 5 | ..... |
| 6 | -.... | 7 | --... |
| 8 | ---.. | 9 | ----. |
Punctuation
| Character | Code | Character | Code |
|---|---|---|---|
| . | .-.-.- | , | --..-- |
| ? | ..--.. | ' | .----. |
| : | ---... | ; | -.-.-. |
| = | -...- | + | .-.-. |
| - | -....- | " | .-..-. |
| ( | -.--. | ) | -.--.- |
| @ | .--.-. |
Two patterns make the chart easier to remember. The most frequent English letters get the shortest codes: E is a single dot and T a single dash. And every digit is exactly five signals long: 1 is one dot followed by four dashes, 2 is two dots then three dashes, and so on up to 5 as five dots; from 6 onward the pattern flips, so 6 is a dash followed by four dots and 0 is five dashes.
Timing rules and what WPM means
Morse code is defined by timing as much as by symbols. Take the length of a dot as one unit. A dash lasts three units, the gap between the signals inside a letter is one unit, the gap between letters is three units, and the gap between words is seven units. Speed in words per minute (WPM) is defined by the standard word "PARIS", which comes to exactly 50 units with these rules, so one unit lasts 1.2 divided by the WPM value, in seconds.
| Speed (WPM) | Dot (1 unit) | Dash (3 units) | One SOS (27 units) |
|---|---|---|---|
| 5 | 240 ms | 720 ms | about 6.48 s |
| 10 | 120 ms | 360 ms | about 3.24 s |
| 15 | 80 ms | 240 ms | about 2.16 s |
| 20 | 60 ms | 180 ms | about 1.62 s |
The sound playback follows these ratios at the speed you choose, with a fixed 600 Hz tone. The speed setting scales the dots, dashes and every gap together.
Worked examples
- SOS becomes
... --- .... Because letters are separated by spaces, the run-together form...---...is one unknown group and will not decode here. Operators often send SOS as a continuous string without letter gaps, so the written form differs from that convention. - Hello World becomes
.... . .-.. .-.. --- / .-- --- .-. .-.. -... Case does not matter. - Help me, please. becomes
.... . .-.. .--. / -- . --..-- / .--. .-.. . .- ... . .-.-.-. Commas and periods are converted too. - Cafe 2024 becomes
-.-. .- ..-. . / ..--- ----- ..--- ....-. If you type an accented letter such as the Γ© in CafΓ©, that character is skipped and listed in the message under the box. - Decoding works in reverse:
.... . .-.. .-.. --- / .-- --- .-. .-.. -..turns back into HELLO WORLD.
Sending words that are not in the English alphabet
This tool covers the Latin alphabet, digits and basic punctuation. For a name or a message in another script, the simplest approach is to spell it with Latin letters first. A name written as "Kim" becomes -.- .. --. The receiver then needs to know which romanization you used, since many languages have several spelling systems, so agree on the spelling in advance. Some languages do have their own Morse extensions or related systems, but they are different codes and are not implemented here.
Tips for learning the code by ear
- Learn letters as sounds ("di-dah", "dah-di-di-dit"), not as counted dots and dashes. Sound-based learning is what most instructors recommend for receiving practice.
- Start at 5 WPM so each letter is clearly separated, then raise the speed while keeping the spacing. The speed menu here changes dot length and all gaps together.
- Learn the six letters that use one or two signals first: E, T, I, A, N and M. Longer letters then start to look like combinations of them.
- Digits follow a regular five-signal pattern, which makes them a good first target for sending dates, times and coordinates.
Limits and cautions
- Only a limited character set is converted. Accented letters and symbols such as !, / and & are not supported, and "/" is used as the word separator.
- Playback is for practice. It needs browser audio to be allowed and is not a radio transmission.
- Morse code is not encryption. The chart is public, so anyone can decode a message. Do not rely on it to keep anything secret.
- Spacing matters. A missing space between letters, or extra spaces, changes how the text is decoded. Always use "/" between words.
Using the result
When memorizing, replaying the converted text as sound at 5 to 10 WPM works better than staring at dots and dashes. If you are building a puzzle, copy the result with the button next to the speed menu. Related tools on this site: Base64 encoder shows another way to represent text with a limited set of symbols, the URL encoder handles percent-encoding of special characters, the text case converter normalizes capitalization before you convert, and the word counter helps you size a message.
Frequently Asked Questions
Why is SOS the distress signal?
SOS (... --- ...) is three dots, three dashes and three dots, a simple pattern that is hard to mistake. It was adopted as a distress signal in international radiotelegraph agreements in the early 1900s. The idea that it stands for "Save Our Souls" is generally regarded as a later interpretation rather than the origin.
Is Morse code still used?
It is rare in commercial communication, but amateur radio operators still use it actively, and it is known to be used for identification signals of some navigation aids.
Why is E one dot while Q is so long?
Codes were assigned so that frequent English letters are short, which saves transmission time. E (.) and T (-) are one signal each, while rarer letters such as Q (--.-) and J (.---) take four.
Why do I get "?" when decoding?
Usually a space between letters is missing (for example ...---...), characters other than dots and dashes are mixed in, or the group is not in the chart. The message under the input box lists the groups that failed.
Why is there no sound?
Browsers only allow audio after you interact with the page, so press the sound button yourself. The tab must not be muted and the device volume must be up. Nothing plays if there is no Morse code in the result.