The History of Speech Recognition – Part 2

Speech recognition has its origins way before computers as we know them existed. Back in the 1880s Alexander Graham Bell invented a way to record speech onto wax cylinders. Others improved on the technique, moving through technologies like plastic and magnetic tape.

When computers became sophisticated enough to include memory and processing capabilities, speech recognition took some giant steps forwards. But early speech recognition was limited to very small vocabularies, and had to be trained with specific voices. They used pattern matching techniques.

A new way of working

When mathematicians Dr Jim Baker and Dr Janet Baker decided to work on speech recognition, they changed the paradigm, using predictive techniques to help with recognition. They launched Dragon Systems in 1982. Several products later, in 1997, came Dragon Naturally Speaking, the first continuous speech recognition software. Users could speak in their natural way without leaving pauses between spoken words.

Still, even at this time, users needed to familiarise the software with their voices, going through a short training regime involving reading texts aloud into the system.

In 2006 came another major breakthrough, with the release of Dragon Naturally Speaking 9. For the first time users did not have to go through an initial enrollment and speech recognition was up to 99 percent accurate right out of the box. Users could literally install the software and start dictating right away.

Expanding technological horizons

In 2007, Dragon celebrated its 10th birthday, and could now boast the ability to recognize 160 words a minute with 99 percent accuracy. Dragon had also gone international, being available in English, Dutch, French, German, Italian and Spanish. And in the previous year special editions of Dragon Naturally Speaking had been introduced for the medical and legal professions, allowing those sectors to benefit from sector-specific language recognition.

These breakthroughs are all the more remarkable when we consider what else was happening in the world of technology in 2007 – it was a long way away from today’s technology landscape:

  • Apple launched the first iPhone (the iPad didn’t come along until 2010, Android made its first appearance in a commercially available phone in 2008)
  • Hitachi introduced the world’s first 1TB hard drive
  • Amazon released its first Kindle in the US (The Kindle became available outside the US in 2009)

In the years between then and now, Dragon has gone on to embrace new technologies as they have been developed. For example, Dragon has utilised Deep Learning techniques to improve its ability to recognise individual’s speech patterns, improving its predictive ability and thus its speed and its accuracy.

Meanwhile the rise and rise of smartphones has opened the way for a new Dragon application – Dragon Anywhere – which was launched in 2015. Dragon Anywhere runs on Android and iOS. It can access personalised standard phrases and commands, using cloud to synchronize with Dragon back at the office. This means users can start a dictated document on the move, and finish it back at the office.

In those early days of the 1980s smartphones, cloud computing, artificial intelligence and other aspects of computing that we take for granted today did not exist. Dragon has embraced them all in pursuit of its mission to increase documentation productivity and improve work processes so that customers can save time and money, improve customer service and work towards their own goals for success.

Learn more

Learn more about Dragon

Learn more

Let’s build what’s next together

Find out what Nuance can do for your business.

Contact Us
Alixson Bell

About Alixson Bell

Alixson graduated from the University of Washington in 2008 and promptly started her career in the computer software industry at a company called Varolii Corporation in Seattle, WA. After Varolii was acquired by Nuance, she quickly became enamored by the many ways to improve customer satisfaction through omni-channel customer service solutions, all powered by AI. She is now a Senior marketing project manager for Nuance, but more than that, she is a firm believer in Nuance’s technology and teams who are tirelessly focused on customer-centric outcomes and results.