Voice Recognition Frequently Asked Questions!

As you can imagine, we're contacted on a daily basis about using voice recognition technology as a blind person. Here are answers to some of the most Frequently Asked Questions. Do check this page often as we add more of them.

Q. When I start Dragon, I get a message to say "an internal recogniser error has occurred". The computer cannot be controlled by voice. What can I do?

A. This is by far the most frequently asked question. It usually results in the computer not being closed down correctly. This may not be the fault of the user. It could be that someone has inadvertently powered the machine off without Dragon NaturallySpeaking being closed down.

We recommend that you backup your speech files once a week. Dragon learns from you all the time as you speak to it, so backing up your profile is helpful. In the event it no longer functions, restoring it returns you to a working situation quickly and with fully developed accuracy. Be sure to back up your profile, (or encourage your student to do it), once a week. It takes all of 15 to 20 seconds!

Using J-Say, say, "Start Backup". A similar utility exists in J-Dictate under help, ALT+Windows+H.

To close the computer down in the correct manner when using J-Say, say, "Computer Off". The speech files are then saved, and the computer is closed down, in the proper order.

If using the keyboard, you should press Control+Shift+F12 to enter the Dragon menu, press Up Arrow once to reach the option entitled "Exit", then press Enter. The computer can then be closed down as usual.

Q. What is the difference between J-Dictate and J-Say?

A. Good question. If you've ever wanted to dictate text into the computer as it is faster than you can type, but you are quite content to edit your text with the keyboard, J-Dictate, (a low-cost product), is for you. It will echo back your dictated text, adjusts Dragon NaturallySpeaking in many ways to make it more accessible, ensures the correction system is usable with JAWS, delivers assistance to support transcription from a portable recorder, and contains other useful features too. It works with all versions of Dragon: Home, Premium, Professional, Professional Individual, Medical and Legal. J-Dictate functions with JAWS for Windows version 11 through to 2019. Our customers of this product include lawyers, social workers, authors of books and plays, and health professionals.

J-Say on the other hand provides complete hands-free control of the computer. So whether you want to save a file, open a folder with a single voice command, browse the internet, manage your Email, listen to the radio etc, all of those things and much much more can be accomplished with your voice. Customers of this product include people with repetitive strain injury (RSI), professionals in work who find it difficult to use the keyboard for whatever reason, those who have suffered a stroke or similar illness, or people who are confined to a wheelchair or bed requiring 24 hour care. If you live in the UK and require training, we have a great deal of experience working with people who have a combination of visual impairment and severe physical disability.

With both products, many changes are made to Dragon NaturallySpeaking to ensure it is accessible to screen-reader users and gives the best performance when working with it. This is done without intervention from the user.
J-Say requires Dragon NaturallySpeaking Professional and JAWS for Windows 17.0 through to 2019.

Q. I want to control the computer with my voice but sometimes I would like to use the keyboard. Can I do that?

A. Indeed yes. When the microphone is disengaged, JAWS behaves in exactly the same way as would usually be the case. No, forget that. It is perhaps better than usual because you've got lots of special functions at your disposal, all of which can be used with the keyboard as well as by voice. Keystrokes and voice commands are given in our extensive product documentation.

Q. Why do I need Dragon at all? I thought Windows had speech recognition built into it?

A. It does. Speech recognition was introduced when Windows Vista was launched. Very shortly afterwards, we produced a product called J-Vist, which did a good job of providing access to it, including echo back of dictated text, access to the Correction System, and context sensitive help so as to assist the user to create a good speech model.

However, we would describe the voice recognition accuracy as, "Adequate". We wouldn't recommend it for use in a professional capacity or if you were totally reliant upon it for daily life tasks, even with the best microphones that money can buy. It's a little like moving into a new home. You're probably not going to be particularly enamoured with the decorating carried out by the previous tennants, or the carpets and curtains they've left behind. You want to get something new and much better. That's the case here. Nuance are the world leaders in voice recognition, and if you need good accuracy, Dragon NaturallySpeaking really is the way to go. Windows Narrator provides adequate screen-reading, but if you could, you'd probably install an alternative screen-reader fairly quickly. Windows voice recognition is the equivalent.

To be more scientific, we dictated a passage comprising 300 words and straightforward-, easy-to-understand vocabulary. Windows Speech Recognition produced a 93 percent accuracy rate, while Dragon NaturallySpeaking gave us 99 percent. We used a variety of noise cancelling headset microphones for the testing, together with a Heil PR40 studio microphone, with consistent results.

Q. If Vispero release an update to JAWS, will that affect J-Dictate or J-Say?

A. Absolutely not. Vispero are being quite prolific at the moment in releasing JAWS updates. You can safely install these updates and your dictation product will still work fine. J-Say supports all versions of JAWS from 17 to 2022. J-Dictate supports from versions 11 through to 2022.
Please note: when we refer to "updates", we do not mean "upgrades" of JAWS. If you move from a major JAWS version 2021 to 2022 for example, that is an "upgrade" and as such you will need a new release of J-Say for that purpose. The J-Dictate product should not require a paid for upgrade.

Q. Can I dictate into most applications and receive echo back of my dictation?

A. Yes. It is very rare these days to find a program which Dragon will not allow you to dictate into. When you speak your text, it will be echoed back to you with our products.

In the event dictating into a program is not possible, Dragon provides a feature called the Dictation Box. This is a straightforward edit box into which you can dictate. When the "OK" Button is activated, the text is "scooped up" from the Edit Field and "dumped" into the program where the cursor is located. But by and large, this should not be necessary.

Q. How do I control an application with J-Say?

A. Theoretically, you can control many programs by saying the word, "Press", followed by its keystroke equivalent. So if you know keyboard shortcuts, (with the exception of those used for JAWS screen-reading functions), you're onto something.

However, with J-say, we believe in creating:

1. A library of intuitive voice commands so you can say what you would like to happen. So with Facebook as an example, you can say, "Go to Messages", and with iTunes you can speak the command, "Get iTunes Store".
2. Behind many of these commands is "special code" so as to suppress unwanted speech output from the screen-reader. As a basic example, in Microsoft Excel, each time you dictate into a cell, you would hear the term "Rich Edit Control". This is annoying for the advanced user and disconcerting for the computer beginner. Many people who use our technologies are fearful of computers because they've never used them. They're not interested in such terms and they shouldn't have to be confronted with them. We take good care to eliminate such speech output when we can to provide our users with a good experience, as we have in our support for Excel.

Q. Can I use a standard headset/microphone in wich to speak to control my computer?

A. Theoretically yes, although we do not recommend it. We would suggest you talk to us or your J-Say/J-Dictate reseller about the best microphone to use in your situation. Are you working in an open plan office? Do you prefer the JAWS speech to come through the computer speakers rather than the headset? Do you prefer two earphones or one? Are you a person who has no hands and therefore cannot put a headset on your head? Many factors need to be taken into account, not only when choosing a suitable microphone, but also when considering using a digital recorder for transcribing your dictated notes. If you are blind, there are hardware considerations as well as determining whether its features are suitable. Do talk to us about any of these requirements prior to purchase.

Q. Is dictating text really faster than typing?

A. Yes it is. For some people of course, it is the only chance they are ever going to have of benefiting from the wonderful technology keyboard users take for granted, and if that is the case, then its a know-brainer.

If you do have keyboarding ability, it is still faster than typing. However, adopting a good dictation technique takes discipline and responding to advice. You can find lots of information about effective voice usage in our product documentation. However, feel free to ask your J-Say/J-Dictate distributor about training. Someone who is experienced in this field, sitting with you to assist you initially and commenting on your performance thereafter, really pays dividends.

As a side note, many people who come to our technologies for the first time have never used a computer. In addition to showing you how to get the most from working with your computer by voice, a good Trainor will spend time helping you to develop your listening skills, so you can appreciate what is being spoken and how to respond when specific events occur. If you do not know why your screen-reader is saying something, you won't know what to do!

Q. With J-say installed, how can I adjust the JAWS Typing Echo and computer's System Volume?

A. These are certainly two things users may like to do, or Trainers may wish to work through this procedure when assisting a new computer user.

To permanently adjust the text JAWS speaks when typing, press Insert+2 repeatedly. Each time this is done, JAWS rotates between the settings of characters, words, characters and words, and, none, representing no audible feedback. This setting is retained even if the computer is restarted.

To achieve the same by voice, say "Keyboard Characters", "Keyboard Words", "Keyboard Characters and Words", and, "Keyboard None".

When you are speaking to the computer using Dragon and J-Say, (having activated the microphone with a voice command), the keyboard echo is automatically set to None, meaning that no characters are echoed when typing. This is perfectly normal. If this is not done, as you navigate some applications JAWS emits erroneous keyboard characters. This should not matter anyway, because you should not be using the keyboard when the microphone is engaged. It should only be activated when you are actually talking to the computer - a fact you will have learned if you receive good quality training.

When the microphone is "put to sleep" by voice, the typing echo setting reverts to the setting which was active prior to activating the microphone. So if you had set it to characters by pressing Insert+2, that is the setting you will be using once again.

To adjust the volume, particularly the speech output, press the Control+Windows+Shift+Up and Down Arrow keys. JAWS announces the volume percentage reached.

To achieve the same by voice, say "Speak Louder", and, "Speak Softer".

Q. Sometimes JAWS announces colour information when reading text. How can I stop this from happening?

A. J-Say contains a number of commands to adjust the speech output from JAWS, such as announcing formatting and colour-based information. One of these functions may have inadvertently been activated by voice.

To reset the speech output, say, "Formatting Off".

Q. If I purchase a product, how do I learn how to use it and can I get problems solved?

A. Given that many people are completely reliant upon voice recognition technology, they may require much more assistance that someone using the keyboard. We pride ourselves therefore on the thoroughness of our documentation. In the case of J-Say particularly, there are hundreds of pages of tutorial information, broken down into "bite-size" chunks for easy reference and learning. Documentation can be provided in DAISY format if necessary.

Technical Support is available at no cost to the user. We do suggest that if a product has been purchased from a local distributor, Technical Support should be obtained from that company, particularly if the problem is of an urgent nature, as the time difference between the UK and other countries can be considerable. However, we are here to help if we can.

Q. Many of my friends enjoy social media. If I use J-Say, can I be part of that as well?

A. Yes you can. We already have full control of Skype by voice, so as to manage audio calls, and being able to hear (and individually review) incoming text chat messages. Support for Facebook also exists. Within the next update, we hope to make it possible to interact with Twitter as well.

Q. I am a low vision user. Can your products help me?

A. If you are using Zoomtext Fusion, certainly J-Say or J-Dictate can be used in conjunction with it without a problem. As well as voice-based equivalents to the hundreds of commands which exist within JAWS, J-Say now contains a variety of voice commands to control the visual elements of ZoomText Fusion. Many of the customers we train are in point of fact using ZoomText Fusion.

Q. Where can I find out more about J-Dictate and J-Say?

A. Please do visit the J-Dictate and J-Say pages of our website for more information.
You can also read about What's New in J-Say 16.0.