We managed to move sound waves vast distances with the telephone, but what if we could get technology to understand and transcribe the sounds for us? This idea brings us to the 1950s.
Step One: Testing… 1…2…3
Computers hadn’t been around for long before engineers were already trying to get computers to listen to us. In 1952, Bell Laboratories invented a program that could understand spoken numbers. Ten years later, at the World Fair, another invention could understand both numbers and simple words (essentially baby talk).
Step Two: Military Intervention
The first significant advances made in the voice recognition software were made by the United States Department of Defense in the 1970s. The part of the military that worked on the project, DARPA, or Defense Advanced Research Projects Agency, is responsible for many technologies we use today, most notably, the internet. DARPA managed to create a voice recognition software called ‘Harpy’ that understood 1,000 words, moving from baby talk to the vocabulary of a 3-year-old.
Step Three: More than Gibberish
The next step came with getting the program to realize that words not in its knowledge base might be words and not just sounds. This step helped the program to move beyond sound recognition and into understanding language; however, every word needed to be said separately so the program could comprehend each individual word.
Step Four: Getting People Talking
With the 1990s came faster computer processors. The processing speeds meant that users of voice recognition software could now speak relatively normally and the program could understand, for the most part, what the user said. The most famous voice recognition software, Dragon, first came out in 1990. Although it brought voice recognition to the masses, it was expensive and only had an 80% accuracy rating.
Step Five: Striving for Perfection
The turn of the century brought no new changes to the voice recognition industry. Accuracy was stuck at 80%. Then came Google.
Google, like Amazon, is an Internet giant that keeps expanding its field of services. Google brought voice recognition software to the next level by creating an app that allows you to search the internet simply by using your voice. This step expanded the words the program could recognize and quickened its processing power with cloud-based data centers. This advance helped programs like Dragon to reach the high 90s for accuracy percentage.
The process of reciting numbers, to talking, to using Siri has taken time and effort from multiple sources. As it continues to develop, voice recognition is a technology that will bring us to a bright, new future.
Rachel Molitor, Writer and Editor, Asta Publications
Asta Publications has a long history of helping writers tell their stories and get published. Since 2004, Asta Publications has helped hundreds of authors bring their book concepts to life and we are ready to help you too! Our dedication to our authors is unmatched. We deliver first-class products and services that are accurate, high quality, and exceed our authors’ expectations.