Text to Morse & Binary Converter

Morse is not really a set of glyphs — it is a rhythm. A dash lasts three times a dot, the silence inside a letter lasts one dot, between letters three, between words seven, and an operator hears all of that rather than reading it. So this converter does not stop at printing dots and dashes: press Play and it keys a sidetone through your speakers on that exact timing, at a speed you set in words per minute, with a lamp that blinks along and a WAV file you can keep.

Two modes share the same pair of boxes. Text ↔ Morse handles the international alphabet — 26 letters, 10 digits, 18 punctuation marks — plus the accented and national letters of the ITU extension and the prosigns operators actually use, written between angle brackets as <SOS> or <AR>. Text ↔ Binary treats a character as the number it is stored as, in base 2, 8, 10 or 16. Type on either side: whichever box you are typing in is marked as the source, and the other one follows it.

The rule throughout is that nothing disappears quietly. If a character has no Morse code — a euro sign, a hash, an em dash — the tool names it instead of deleting it, and offers to fold accented letters down to plain ones so the rest of the sentence can be sent. If a run of binary is not a whole number of bytes, that is an error with the offending group quoted back to you, not a wrong character presented as a right one.

Speed is quoted in words per minute against a fixed reference word — PARIS followed by a space, chosen because it happens to measure exactly 50 units of dots, dashes and silence. One dot therefore lasts 1200 divided by the speed, in milliseconds: 60 ms at 20 wpm, 240 ms at a beginner's 5 wpm. The slider here covers 5 to 40 wpm and recomputes every duration from that single figure.

The second slider is what makes the tool useful for learning rather than only for listening. Sending everything slowly teaches you to count dots, which is a habit you then have to unlearn, because a letter at speed is a shape rather than a sequence. Farnsworth spacing solves it by sending the characters at full speed and stretching only the silence between them, so the shapes are the ones you will eventually hear while the thinking time is generous. Set the character speed to 18 and the overall speed to 5 and the gaps take up the difference by the standard formula, holding the reference word at twelve seconds.

Pitch matters more than it sounds like it should. A tone that sits clear of the noise your ear is already filtering is far easier to follow, which is why receivers settled on something in the 600 Hz region; the slider runs from 400 to 1000 Hz. Each element also gets a five-millisecond fade in and out, because a sine wave switched on and off abruptly produces a click at every edge that is louder and more tiring than the tone itself.

The international alphabet is small on purpose, and the gaps in it are real. There is no code for a hash, a percent sign, an asterisk, a tilde, a bracket pair beyond round ones, or a caret — fourteen printable keyboard characters in all. Accented letters are better served: the ITU extension and national practice between them cover à ä ç é è ñ ö ø ü ż and a couple of dozen more, all in the table here, and several deliberately share a code because Morse has fewer distinct signals than Europe has letters. Ä, Æ and Ą are one signal; so are Ó, Ö and Ø. Prosigns share codes with punctuation for the same reason, which is why the chart lists what each code decodes back to: the plus sign and the prosign AR are the same thing being read two ways.

Reading Morse back is deliberately forgiving. Charts are typeset with a middle dot and a Unicode minus rather than a full stop and a hyphen, some sources separate words with a pipe or a double slash, and pasted text arrives with all of it mixed. All of those are accepted, and a group that still cannot be read is quoted back with its position rather than swallowed.

The numeric side answers a different question, and the default answer is UTF-8 bytes. That distinction is worth a moment: a character's code point is its number in Unicode, but what is stored on disk is a sequence of bytes encoding that number, and only for the first 128 characters are the two the same thing. An emoji is one code point and four bytes; the euro sign is one code point and three. Grouping by byte is what makes "eight bits each" a true statement and what lets any character survive the trip out and back. Code points and UTF-16 units are both available too, labelled for what they are, and the inspector puts all of them in one table next to the character's Morse.