Visual sound traces cannot uniquely capture what listeners perceive as the same vowel

Evidence

Alexander Graham Bell, a teacher of the deaf like his father before him, used the phonautograph to train his students to visualize the sound they could not hear. Many students could not speak understandably; since birth they had lacked auditory feedback from their own voices. Bell’s idea was to train them to reproduce phonautograph signatures recorded from ordinary speech. If the signature looked right, it would sound right! This is actually true; however, it is too restrictive—many sounds with different-looking sound traces are perceptually nearly identical. Worse, the normal variations in physiology might make the sound trace for the vowel sound “ohhhh” in subject A impossible for subject B to reproduce, yet both are recognizable as the same vowel. The concept Bell really needed was years in the future: formants, which are broad ranges of enhanced frequencies in the voice that define different vowels (see chapter 17). — Eric J. Heller, Why You Hear What You Hear- An…, 465

Categorization

• DEFINE: Introduces formants as broad ranges of enhanced voice frequencies defining different vowels, though detail deferred to chapter 17.
• XREF: The point that distinct physical signals map to identical perception foreshadows the book’s wider theme that perception is not a literal transcription of the waveform.