From visible speech to voiceprints - blessing and curse of sound spectrography

Abstract

: Sound spectrography is probably the single most widely used tool in acoustic phonetics today. It was first called Visible Speech, thus linking it to Alexander Melville Bell's 1867 publication. At that time, of course, the term only referred to auditory analysis and the iconic documentation thereof. During World War II, sound spectrography as we know it today was developed in the US as part of a war project and was therefore not published until well after the war. A similar device was built in the Soviet Union in the post-war period. Although claims were made on both sides that the device was intended for the benefit of the congenitally deaf, it is quite clear that its primary purpose was in speaker identification. This contribution focuses on the development of the sound spectrograph and outlines its role in (forensic) speaker identification.