Music Steganography and Musical Cryptograms

Music offers many interesting ways of hiding messages. An early example is, Tre rebus musicali, written by Leonardo da Vinci. Many great classical composers have included musical cryptograms, often their name, in their works. Nowadays, audio steganography tools can be used to hide messages, or any file, inside of an audio file.

Boxentriq puzzle promotion

Audio steganography

Audio steganography makes subtle changes to the music or audio that we can't perceive with our ear. For instance, we can't hear frequencies lower than about 20 Hz or higher than 20 kHz. There are examples of shady advertising companies listening to the microphone on mobile phones for ultrasonic audio beacons, to track the behavior of the users.

A simple way of hiding information in audio files would be to use the LSB (Least Significant Bit). The noise it will cause is usually so small that it will be impossible to hear. To further protect it from discovery, it could be applied to every 2:nd, 3:rd or n:th sample instead of on consequtive samples. Error correction algorithms can be used to improve this method further. More advanced audio steganography methods hide secret information in inaudible parts of the frequency spectrum.

Audio steganography tools

How to identify music

There are some tools available that can listen to any music through a microphone or a audio file, and try to identify it for you. Shazam is one of the most popular tools. Identifying an unknown piece of music can be very helpful to find out about any modifications done.

Music identification tools

Musical cryptograms

In a musical cryptogram (also called music cryptogram), the composer has chosen notes so that they can be translated to a message. One of the most famous composers, Johann Sebastian Bach, included the notes B-A-C-H in many of his works. Creating a musical cryptogram is not too difficult, but the great composers can make it sound natural and enjoyable to listen to. Basically in a musical cryptogram, there are two elements that can be used: tone pitch and tone length. The earliest, and easiest, method is to transcribe tones to their names. Below is a table with some common transcriptions:

Tone transcription table
Frequency Common name (English) Germanic variant Solfège (original) Solfège (modern)
261.63 C C Ut Do
293.66 D D Re Re
329.63 E E Mi Mi
349.23 F F Fa Fa
392.00 G G Sol Sol
440.00 A A La La
493.88 B H Si Ti
The French method

A popular method among French composers during the 19th century was to arrange letters in a table. Any letter was then represented by the note in the top of the column. For instance, D, K, R or Y were represented by the note D. This many-to-one mapping method had the disadvantage that it is difficult to extract the original message (and sometimes the message can be ambigious). To avoid that problem, the notes could be altered, such as being played in different octaves or of different lengths.


Using pairs of notes, the French method can be enhanced so that one of the notes describes the row and the other note describes the column. This is similar to a Polybius square, used in classical ciphers. Musical cryptograms are therefore similar to classical ciphers.

Note lengths

The note lengths also play a key part in some kind of musical cryptograms. The famous 20th-century composer Olivier Messiaen developed his own musical cipher, involving both tone pitch and tone length, and used it in his organ works Méditations sur le Mystère de la Sainte Trinité.

Other possibilites include using morse code or similar schemes to transcribe the message to note lengths.


Another technique of including hidden messages in music is to record them backward onto the track. The technique was used by some black metal bands and musical artists to include extra lyrics in their songs. One of the earliest examples of backmasking was The Beatles including "...the sun shines. Raaain. If the rain comes, they run and hide their heads" in their song "Rain". This backmasking was later acknowledged by John Lennon.

Backmasking can easily be restored by reversing the sound in any audio editor.

Audio editors