The human voice is produced by the complex interaction of breath, vocal folds, resonating chambers (chest, throat, sinuses, skull) and the articulators (tongue, lips, palate). Every aspect of this system is affected by the speaker's physiological and psychological state — making the voice an extraordinarily rich channel of information that operates largely outside conscious control.
Research in voice science has established that listeners make surprisingly accurate judgments about speakers from brief voice samples. Studies have found that people can assess dominance, trustworthiness, physical size, health, age and emotional state from voice alone with accuracy significantly above chance — and that these judgments happen within the first few hundred milliseconds of hearing someone speak. The voice is processed rapidly and largely unconsciously; the impressions it creates precede any conscious analysis of content.
The evolutionary logic is clear: before language became the primary channel of communication, vocal quality was the primary signal for assessing the physical and psychological condition of other individuals. The systems for processing these signals are ancient and automatic — which is why they are so difficult to fake reliably and why trained actors spend years learning to do so.