Dragon Naturally Speaking 8

December 14th, 2004 Software, Tech, The West Australian

Most of us know there’s such a thing as voice recognition in computers. We’ve all seen Hollywood movies where a high ranking general gains entry into the top secret bunker beneath a mountain by speaking into a microphone and having the computer scan the timbre of his voice to confirm it’s really him.

Who could forget Harrison Ford as Lt Decker in Ridley Scott’s seminal 1982 film Bladerunner? In one sequence, the grizzled detective is looking over security camera footage searching for evidence. He pans from one side of the image to the other, zooms in, zooms out, advances and retraces through the film, giving all his commands by voice.

We all assume that’s the future of voce recognition technology — interaction with a computer completely by voice with no more need for squinting or RSI. But here’s a shock; we’re at that stage now.

If you haven’t had much experience with voice recognition, you probably think as most of the uninitiated do; that it means having to speak in a monotone, leaving huge pauses between words, even leaving the inflections of your accent out of your voice as much as possible.

And even then, you probably have to go back through to correct the innumerable mistakes caused by your accent, the pitch or volume in your voice or by not sounding like a computer yourself.

The setup of Dragon Naturally Speaking gives you some idea what’s in store. Part of the installation process is choosing a local accent that you’ll be speaking in, which seems very promising.

Them it runs you through a series of quick tests so it knows your voice and gets more used to it — one read-through lets it gauge the tones of your normal speaking voice, another the volume.

But it doesn’t just get itself used to you, it gets itself used to your work. One of the first tasks you can do with it is assign it to scan a collection of documents — every one on your disk if necessary.

It’ll compile a list of unrecognised words from documents it’s scanned, and you have only to agree to or dismiss them one by one. Why? Say you work in mining and your documents are full of words like ‘Rio Tinto’, ‘Mt Hammersley’ and ‘bauxite’. You don’t want it getting those words wrong — just choose it from the list, say it, and next time you say it (presumably often), Naturally Speaking will remember it.

And if that fails and you use more words not found in too many dictionaries (your name — for instance — along with many other proper nouns), you simply add it to your dictionary and it’s there for good — recognised every time.

Something a lot of people also think about voice recognition software is that it’s restricted to the major portions of your workload only within your applications. You know — you start a new word processor document by hand (ie with the mouse) and speak your letter or memo, saving or printing manually before moving onto the next task via the keyboard or mouse again.

You’re in for a very pleasant surprise. You set a single switch that toggles the program on or off, or alternately you give it a voice command to wake up or go to sleep, letting you work as you normally would while it dozes in the background.

But from the time you activate the program, you can do simply everything by voice. Sitting there saying things like ‘start Microsoft Word’, ‘switch to Outlook’, ‘start control panel’, ‘new message’ and ‘minimise that’ will make you feel so futuristic you’ll be expecting to teleport to work or go to the shops in your flying car.

There are of course times where it’s quicker to just enact a few clicks with the usual input apparatus, and for that the toggle key is invaluable. Naturally Speaking isn’t about making you do everything by voice, it’s about letting you use voice commands where necessary to increase your productivity (or ease your typing fingers).

And the array of commands already built in and ready to use in your usual applications is staggering. Theoretically (as an example), you can leave the keyboard untouched to not only write but also format a whole word processing document. Speaking commands to select text anywhere in the document, change the size or colour, replace it with another word or add it to the dictionary on the fly is a very surreal experience if you’ve never done it before.

It would take a long time to work your way completely through the program and have it 100% accurate to your situation and voice (in fact, it’s designed to learn and grow with you), there’s just too much flexibility. For that reason, it’s as if Dragon Naturally Speaking was made especially for you.


Full client and publication list:

  • APC
  • AskMen.com
  • Auscam
  • Australian Creative
  • Australian Macworld
  • Australian Way (Qantas)
  • Big Issue
  • Black Velvet Seductions
  • Black+White
  • Bookseller & Publisher
  • Box Magazine
  • Brain World
  • Business News
  • Business NSW
  • Campaign Brief
  • Capture
  • CHUD.com
  • Cleo
  • Cosmos
  • Cream
  • Curve
  • Daily Telegraph
  • Dark Horizons
  • Dazed and Confused
  • Desktop
  • DG
  • Digital Media
  • Disney Magazine
  • DNA Magazine
  • Empire
  • Empty Magazine
  • Famous Monsters of Filmland
  • Fast Thinking
  • FHM UK
  • Filmink
  • Follow Gentlemen
  • Geek Magazine
  • Good Reading
  • Good Weekend
  • GQ
  • How It Works
  • Hydrapinion
  • Inside Film
  • Internet.au
  • Loaded
  • M2 Magazine
  • Marie Claire Australia
  • Marketing
  • Maxim Australia
  • Men's Style
  • Metro
  • Moviehole
  • MSN
  • Nine To Five
  • Paranormal
  • PC Authority
  • PC Powerplay
  • PC Update
  • PC User
  • PC World
  • Penthouse
  • People
  • Pixelmag
  • Popular Science
  • Post Magazine
  • Ralph
  • Reader's Digest
  • ScienceNetwork WA
  • SciFiNow
  • Scoop
  • Scoop Traveller
  • Seaside Observer
  • SFX
  • Sydney Morning Herald
  • The Australian
  • The Retiree
  • The Sun Herald
  • The West Australian
  • thevine.com.au
  • TimeOut
  • Total Film
  • Video Camera
  • Video&Filmmaker
  • Writing Magazine
  • Xpress
  • Zoo