|
speech recognition |
 |
 |
Frequently Asked Questions
- What is continuous speech recognition?
- What is the difference between the Dragon NaturallySpeaking Preferred and Dragon NaturallySpeaking Professional Editions?
- What are the minimum system requirements for Dragon NaturallySpeaking?
- Will I need formal training for Dragon NaturallySpeaking?
- I used speech-recognition software a few years ago and had poor results. Has the accuracy improved?
- How long does it take to ‘train’ my PC to recognise my voice?
q.
What is continuous speech recognition?
Continuous speech recognition is when you can speak at the same pace as most
of your conversations and your spoken words are transcribed immediately onto
your computer screen.
Back to the top of this page>
Dragon NaturallySpeaking uses continuous speech, or natural speech. This eliminates the need to insert a pause between every word. Just talk to your PC naturally, at a conversational pace. Your sentences appear immediately on your screen and in your document. Dictate entire paragraphs at a time. Compose e-mail messages, create reports, draft letters and edit documents just by speaking. Using Dragon NaturallySpeaking you can dictate documents at up to a theoretical speed of 160 words per minute.
q.
What is the difference between the Dragon NaturallySpeaking Preferred
and Dragon NaturallySpeaking Professional Editions?
Dragon NaturallySpeaking Professional allows you to create your own customised
scripts known as voice commands. These commands may be simple dictation shortcuts,
such as inserting a paragraph of text into your document which you use frequently,
or more complex scripting procedures. These advanced macros can be customised
to control and navigate specific events within any application, tasks can also
be automated service to sequence of actions can be repeated in a certain order
by saying a simple command. For example, consider the steps in composing an
e-mail to your colleague John:
- Start Microsoft Outlook
- Create a New Mail Message
- Type the Recipients E-mail Address ‘john@mybusiness.com'
- Click in the ‘Subject’ Field ready type a subject
A voice command could be created called ‘New Message to John’ which when issued would perform the above steps without the need to use the keyboard or mouse. This is a very simple example. Any series of steps in an Application which you perform on a regular basis could be automated in this way. For this reason BDF Solutions particularly recommend the use of the NaturallySpeaking Professional Edition for clients wishing to minimise mouse and keyboard use for controlling their PC.
The NaturallySpeaking Preferred Edition is limited to inserting "text and graphics" dictation shortcuts only.
Both the Professional and Preferred editions offer
the same standard of recognition accuracy.
Back to the top of this page>
q. What are
the minimum system requirements for Dragon NaturallySpeaking?>
Dragon NaturallySpeaking products will run satisfactorily on a PIII with 1GHz
processor speed (or equivalent processors) with a minimum of 512MB RAM. For
optimum performance in high demand environments such as medicine, law and at
the corporate level a P4 (or equivalent processors) with 1024MB of RAM is desirable.
You will require a Creative Labs Sound Blaster 16 or equivalent sound card supporting 16-bit recording. ‘On-board’ soundcards where the sound circuitry is integrated into the motherboard of the PC should be avoided. BDF Solutions particularly recommend the use of a USB microphone to eliminate the need for a sound card and ensure a high level of audio input to the PC.
If you are unsure about your particular PC specification
please contact us for advice.
Back to the top of this page>
q. Will
I need training for Dragon NaturallySpeaking?
BDF Solutions recommend training in order to maximise your investment.
BDF Solutions are a Nuance Premier Application Solutions Partner and as such
offer introductory, intermediate and advance training for the NaturallySpeaking
range. We are highly experienced in integrating NaturallySpeaking into workplace
environments and can customise the product for particular user requirements
and business sectors.
Back to the top of this page>
q. I used
speech-recognition software a few years ago and had poor results. Has
the accuracy improved?
Providing you have the correct hardware platform and appropriate training you
could reasonably expect 85-90% accuracy straight away. The key to improving
accuracy lies in approaching corrections in the appropriate way.
Back to the top of this page>
q. How long
does it take to ‘train’ my PC to recognise my voice?
After installing the program and creating your user profile, you only need
to read to the computer for about 5 to 10 minutes. The computer will take additional
time to process your voice. Depending on the speed of your computer, this could
take anywhere from 1-15 minutes.
Back to the top of this page>
Dragon professional  |
|
|
| Dragon preferred |
|
|
| Need our help? |
|
|
Want
more information about the products, services and training we offer?
Complete our contact form and one of our advisors will get back
to you [