by Allana Grant
3 months ago
Voice recognition: an invaluable resource for disabled people
Universally used by banks to identify customers, employed by government agencies to speed up admin processes and built in to smart televisions and cars: software that recognises the human voice and converts it into text, or voice recognition as it is more widely known, is all around us these days. It is one trend that we, as a society, are getting fully on board with; as we increasingly turn to digital assistants such as Siri and Cortana to carry out all manner of tasks in our daily lives.
However, for the millions of people in the UK who have a physical or learning disability and are unable to type, write or use a mouse, this technology represents much more than a simple short cut. It provides an alternative way to operate devices such as computers and smart phones; enabling users to quickly create lengthy text documents and even perform commands using dictation rather than typing. It is an invaluable tool for people with conditions such as dyslexia for example; as words spoken to the computer, more often than not, appear on screen spelled correctly.
How voice recognition software works
Voice recognition programmes work by analysing the sounds a user makes, filtering everything that they say, digitizing it to a format it can “read” and finally extracting meaning from it. Then, working from a combination of algorithms and previous input, it makes an educated guess as to what the user has said which is what appears on screen. Most programmes can be trained to a degree, to recognise the speaker’s use of language and so they become increasingly accurate with the passing of time.
Voice recognition software has come on in leaps and bounds over the last decade especially, and it is improving all the time. As standard, these programmes now enable disabled users to enter text in to a document and edit/format it via a series of simple voice commands such as “delete sentence”. Increasingly however, we are seeing the emergence of more advanced sophisticated products which incorporate commands in to their design to open and run different programs, compose and send emails, and even browse the internet.
A far cry indeed, from Bell Laboratories’ ground-breaking Audrey System. Launched in 1952, the world’s first ever speech recognition technology was a six-foot-high relay rack, containing copious amounts of circuitry, which had the capacity to understand numeric digits spoken in to its speaker box.
Voice recognition solutions
As the technology has advanced down the years; the voice recognition software solutions on offer to disabled users has increased dramatically. These days there are options to meet the most stringent demands as well as ones to suit every budget. There are of course far too many to mention in this post, but let’s look at three of the main players:
It is widely accepted that Nuance’s Dragon Naturally Speaking product range is the market leader when it comes to voice recognition software. It traditionally fairs well on most review sites, scoring in the high nineties in the dictation accuracy steaks. In addition to its main functions for creating and editing documents, it also boasts the most impressive array of features of any software on the market, including the ability to go hands free using a computer’s inbuilt mic, and the option to post to Facebook or Twitter. It is compatible with both Windows and Mac operating systems and can be used on devices across the board.
Although few would argue with Dragon’s performance; as with most assistive technology, it can prove too costly for some,-the Professional edition starts from £279.99. Never fear! There are a plethora of budget or even free, good quality alternatives available on the market though.
Recent years for example, has seen Microsoft and Apple incorporate their own speech recognition technology in to their products.
Every Windows laptop and desktop now has a built-in voice recognition program called Speech Recognition. It may not include as many features as Dragon products, but after creating a voice profile to train the programme, users are able to carry out tasks simply enough using voice commands. Recognition rates are generally reported as being good and users can extend the programme’s functionality through System Macros.
For those who prefer Mac devices; any running OS X Mountain Lion, Mavericks or Yosemite also offer free dictation software. This, it has to be said, is more suited to users with at least some ability to use a keyboard as it is limited to entering, editing and formatting text. There are no options for operating the Mac’s other programmes and so on. Again the speech recognition is fairly reliable, and the latest version of OS X, Yosemite, now includes a range of additional editing and formatting commands, but the programme’s functionality is very much dependent on having a steady internet connection so that recorded speech can be sent to Apple’s servers for analysis and conversion.