Armenian scribes began writing chant signs, called khaz, above the text from about the ninth century, and the system grew steadily more elaborate until the manuscripts of the fourteenth and fifteenth centuries carry dozens of distinct signs. What those signs told a singer to do is no longer known. The notation was not a staff and did not fix pitch. It appears to have cued melodic gestures and ornaments a singer already carried in the ear, and when the oral chain that supplied that knowledge was broken, the signs stopped being readable.
The nineteenth century produced a replacement rather than a decipherment. Hambardzum Limonjian devised a new Armenian notation in the 1810s, borrowing shapes from the khaz signs but assigning them fresh and explicit meanings, and the chant as it was then sung at Constantinople and Etchmiadzin was written down in that system. Nearly all printed Armenian chant descends from those transcriptions, so what is sung today is a nineteenth century record of a living practice rather than a reading of the medieval books.
Komitas Vardapet, the priest and musicologist who did more than anyone to collect and publish Armenian sacred and folk music before 1915, worked on khaz for years and did not solve it. The problem has been taken up repeatedly since, most systematically by Robert At'ayan, without producing an agreed reading. It remains one of the open problems of Eastern Christian musicology, and the atlas records it as open rather than repeating any of the proposed solutions as settled.