The word barytone, which is more familiar to many English speakers as a category of singing voice, actually comes from the Greek language, and also refers to an element of Greek grammar where the last syllable of a word does not get a significant stress. The word is a hybrid of the Greek word "heavy" or "low" bary, and "pitch," i.e. tone. Here, the word "low" is abstracted to refer to a lesser accent or stress for a sound.
In order to understand the barytone in Greek grammar it is first necessary to identify the various parts of a Greek word. These are commonly known as the antepenult, penult, and ultima. These could be otherwise labeled as the first syallable, the second-to-last syllable, and the final -- or ultima -- syllable. The ultima syllable is the syllable that is treated as a barytone. It is commonly understood as a root/abbreviated form of the English word "ultimate," meaning "last."
Many people associate the word baritone with a vocal range for singing that lies between the tenor or higher male register, and the bass or ultra-low register. This use of the original word barytone is more concrete, where the translation to "low" in English would refer directly to the register or pitch of the sound. This is a part of specially structured singing in most societies where overall form relies on the synchronization of a set of voices in various registers. The same word can also be used to talk about the lower ranges of a variety of musical instruments.
The use of barytone in Greek is an example of what's called pitch accent. Pitch accent is an alternative to the concept of stress accent in English. With pitch accent, some syllables are developed for specific meaning or use through a variety in pitch, rather than a variation in stress. Greek is an example of a language that uses pitch accent as part of the function and comprehension of the language.
Within the general category of language using pitch accent, some languages are more reliant in the use of pitch than others. Some languages are sometimes described as "fully tonal," where multiple tones are integral to the use of the language. Some languages use only binary high and low tones, where other languages have a more sophisticated set of three or more tones.