A voice built from small pieces
Animalese is assembled from recorded syllables rather than spoken as a complete performance. The game can join those short audio fragments to produce the sound of a character talking, without needing a voice actor to record every line from beginning to end.
That matters in a game where characters can say many different things. A library of reusable pieces can be combined as dialogue changes, instead of requiring a separate recording for each sentence. The tradeoff is that the result is not ordinary, naturally paced speech: the fragments are played quickly, giving the voices their distinctive babble.
Why it still feels like conversation
Players usually read the dialogue while hearing it. The text carries the precise meaning; Animalese supplies a lively vocal layer that makes an exchange feel spoken without having to deliver every word clearly. Its quick, syllable-by-syllable rhythm also helps keep the sound present without making each conversation depend on a long spoken recording.
The clever part is the division of labor: the screen tells you exactly what was said, while the voice makes the character seem to say it aloud.
