How to give my singing creatures a beautiful voice?

Question

My world contains a race of creature that does not speak to communicate, they sing.

They are humanoids, living in similar conditions as humans. They are intelligent, they have a complex langage, they can cast magic with their voice.

The song have to be a very particular sound that can be sweet or pretty loud. Not only a voice, but several sounds making a beautiful melody that can not be confused with anything else. For comparison I have a church organ in mind for the "song", but I don't know whether it's possible coming from a single humanoid. I have to make one of them mute in the story, what leads to the question :

What kind of organ (tongue, lungs, vocal cords, something totally new...?) does a creature needs to perform such a song ?

Answer 1

The animals in nature with the best singing vices are birds; they achieve their facility by virtue of their syrinx, an organ located at the fork of the trachea.

A humanoid with a syrinx would be able to produce two separate notes simultaneously. It is also shown that some species of bird are able to vary the notes they produce extremely rapidly, potentially altering the note produced with each vibratory cycle of the syrinx membranes, potentially introducing harmonics that the being could use to emulate any sound in their syrinx's pitch range, and could emulate an instrument with more than two notes.

This could quite easily be used to produce sounds that could be considered to sound similar to a pipe organ.

An interesting side effect of this is that said humanoids would simply open their mouths and complex sounds would come out - there would be little need for manipulation of the lips, tongue and pharynx.

The primary disadvantage would be that said humanoids would have difficulty producing sibilant sounds - their tendency would probably be to produce a whistle rather than a hiss - due to the many rapid random variations in syrinx tension that would be required. They should be able to produce sibilants as we do (turbulent airflow past the tongue), but the idea may be quite alien to them initially.

Answer 2

Pipe organs are a very early form of additive synthesis.

Monty Wild's answer alludes to the requirement to produce multiple sounds at once. The solution of using modulation is great for adding harmonics to a single note, but pipe organ music tends to also use chords to produce the characteristic sound.

One way to reduce the number of notes and harmonics needed would be to fix the key of their communication. For instance on a 'western' human keyboard there are 12 semitones that can be combined into a scale. The most musical but sparsest scale I am aware of is a pentatonic scale featuring 5 frequencies in each octave and the numeric doubling and halving of those notes.

As for anatomy to achieve that I'm reaching here. However, you could replace the vocal chords (which have monophonic but continuously variable pitch) with a series of pipes that have fixed tuning, each with the ability to resonate at different harmonics (see the picture at the top of that article to imagine what the sound waves would be doing in those pipes.

From that you would be able to produce a pipe organ like sound with the number of separate tones they can produce being equal to the number of pipes. The higher the fundamental note, the smaller each pipe could be, so you'd probably find that a large portion of their speech may be ultrasonic for human ears.