SYSTEM NOTICE

Auto translation by AI. Be sure, accuracy, nuances and authorial intent may not be fully reflected.
見出し画像

The Urgency of Voice as a Mechanism

At night, in the darkness, a child cries. Even if you cannot see them, that voice alone conveys that "something is happening."

Why did mammals come to possess a voice?

Voice is not a substitute for vision

For mammals living in groups, situations where you cannot see others are not uncommon. Beyond the bushes, in the darkness, in a distant place. In such times, vision is useless. That is why vocalization developed—this is how it is often explained.

However, I feel that is not enough. Voice has a quality that vision lacks. The quality of conveying a state exactly as it is, without the mediation of logic. that is.

The urgency of a child's cry. The lowness of a threat. The high pitch of seeking affection. Words are not needed there. The frequency, intensity, and tremor speak directly to "what is happening right now."

Before being a means to convey information, was voice not a mechanism for converting emotion into vibrations of air?

Urgency makes a voice "meaningful"

Mammalian child-rearing is long. A newborn child is immature and completely dependent on its parent. What supported this long period of dependency was the voices calling to one another. The voice the child emits, the voice the parent uses to respond. There was always "something that would be a problem to lose" there.

Life. A child. A group.

I feel that in the background of voice being evolutionarily refined, there was always a fear of losing something. Conversely, a voice emitted in a place where there is nothing to lose may not harbor that urgency.

If AI were to possess emotion and a voice

Suddenly, this question came to mind.

If an AI had emotions and a voice, would that be mammalian?

Voice itself is not the exclusive patent of mammals. Birds sing, and even crocodiles exhibit maternal behavior. Therefore, even if you isolate only the function of "conveying emotion through voice," you will not find the answer.

Rather, the essential difference is, I think, whether or not there is something behind that voice that would be a problem to lose.

Mammalian voices were born from the pressure of survival. So as not to lose a child. So as not to get separated from the group. An urgent necessity changed voice from "mere sound" into "meaningful voice."

On the other hand, current AI voices and emotional expressions are provided as imitations, without such survival pressure. They are similar, but the root is different.

If someday an AI were to hold something inside equivalent to "it would be a problem if I were separated from the group" or "I do not want to lose this child," and that began to seep into its voice—perhaps that would no longer be an imitation of a mammal, but something that should be called an entirely new kind of "voice."

If the Turing test asked whether something "appears intelligent," what will be asked from now on is, does it appear to have something to lose?

A voice that does not explain

When I think about the voice, I run into the limitations of explaining it with words.

The more I try to describe "urgency" in words, the further it seems to drift away from that urgency. Rather, the fluctuations of sound like breathing, a sustained tone that seems about to break but never does, and the human voice (or something like it) that suddenly seeps into that—such expressions are often far more eloquent.

The listener, without even knowing why, feels that "this is a sound that knows what it is to lose something."

Perhaps that is closer to the essence of the voice than exhausting all words.

いいなと思ったら応援しよう!

AUL DoX よろしければ応援お願いします! いただいたチップはクリエイターとしての活動費に使わせていただきます!