A to Z of Innovations at Expos: Voice recognition

A to Z of Innovations at Expos: Voice recognition

In 2020, voice recognition technology can be used by anyone with a smartphone. Back in the 1960s, the idea of a human operating a machine by using vocal commands seemed like science fiction. Visitors to World Expo 1962 Seattle – the Space Age Expo - were therefore amazed by IBM’s Shoebox – an early computer that responded to voice controls.

Developed by William C. Dersch, the Shoebox was the world’s most advanced speech recognition apparatus, capable of recognising 16 words spoken in English: the ten single digits and six basic commands including addition and subtraction. The innovative device worked by converting voices – spoken into a microphone – into electrical impulses. The computer used these to operate a calculator and print the answers. As with modern speech recognition systems, users could see the words they were saying, ensuring there was no ‘misunderstanding’.

In developing the Shoebox, it was observed that higher or lower pitches were used for different words, leading to the implementation of an audio filter to moderate the tone, so that each word could be correctly distinguished.

Following on from the fascination the technology created at Expo 1962, the development of voice recognition picked up pace in the following decades, but it would not be until well into the 21st century that it became commonplace. Now, far from being limited to simple mathematics, voice recognition is on our mobile phones, computers, watches, cars, and in our homes to help people perform or complete a large number of tasks and respond to queries.

Les points de vue exprimés dans cette rubrique sont ceux de contributeurs externes à l'Organisation et ne reflètent pas nécessairement l'avis et la position du BIE. Consultez cette page pour savoir comment contribuer au blog