Dragon VoiceComputer

I'm adding this topic in order to encourage members of the form to post issues with Dragon as well as Dragon/VoiceComputer. I'm going to be adding to this things that I have discovered that may be helpful to both Dragon users and VoiceComputer users.

Please feel free to post any questions that you have on either and/or both. Will do our best to try to answer them as well as continually update this topic.

Here’s a tip when working with ordinals (i.e., 1st, 32nd, etc.).

Rather than go through all the trouble and hassle of adding ordinals in the Vocabulary Editor, preceding number that you want displayed as an ordinal by saying “numeral” followed by the number that you want displayed as an ordinal. This procedure works all the way from 1st up to 99th. Under some conditions it can be tricky if you’re dictating too fast, but it works. It does not seem to work for ordinals after 99th, but it works fine up to that point.

Interestingly enough, it also works with dates very effectively.

For example, if you want 31st of December without wrapping your leg around your neck, just say “numeral thirty first of December”.

You’ll get what you want and you won’t have to pause. That is, you can dictate that phrase continuously without pausing and you will get the right results. Saves a lot of time and a lot of hassles.

Chucker2017-01-09 11:06:10

I’m new to VoiceComputer, Dragon, and forums, so excuse my missteps. This is my fourth day using voice computer and Dragon, and I really like them in general.

However, for some reason, primarily Microsoft Word but sometimes other programs as well (Microsoft Excel, Lotus Notes), randomly goes into “not responding” mode for 10-15 minutes. During most of that time, I can’t use either VoiceComputer or Dragon commands anywhere (they essentially don’t respond either)In a rate.

Saving a simple Microsoft Word document also takes that long EVERY SINGLE TIME if VoiceComputer and Dragon are open.

The first or second time this happened, I tried to open another version of Word while the first was saving, and I got an error about the speech add-ins, which gave me the clue that the “not responding” problem was related to VoiceComputer and Dragon.

I have restarted both my computer and speech, but no luck.

Any ideas how I can speed this up?
<edited><editID>jlevine</editID><editDate>2017-01-12 11:33:15</editDate></edited>


There are a number of reasons why this may be occurring. However, there are a couple of questions that you need to answer in order to give you a more complete response.

1. What version of Dragon are you using? For the moment I will assume that it’s DPI 15 (Dragon Professional Individual 15).
2. What is your hardware configuration? That is, CPU make and model, total amount of RAM installed and what version of Windows (i.e., Windows 7, Windows 10, 32-bit or 64-bit).
3. What version of Microsoft Office are you using?

As I said, assuming that you’re using DPI 15 open the Dragon Options (i.e., say “show Dragon options”) and go to the Data option. By default, DPI 15 is set up with the “Save recorded dictation with document” set to Always. That setting will store not only the text of your document but also the audio behind as a separate file with your document file name and the extension “dra” (i.e., my document.dra where the name of your actual text document is my document.docs or docx). If that’s the setting that you have, change it to “Never”. Keep in mind that the audio portions of Word document saved with your text will take up 1.5 MB per minute of dictation. If you have a large document that you dictating into for a long time and (i.e., a couple of hours give or take), the dra portion can be 20 to 30 MB or more. Every time you want to save a document under this condition, it will take longer to save, sometimes by a significant amount of time. It will also take longer to load the document. Note that if you have that option set as “Always” you can’t simply set the option to “Never” and expect the latency to go away with documents that you have created under the “Always”. You have to manually go in to where those documents are saved and delete all the corresponding DRA files associated with those documents. Otherwise, every time you load and save the document is going to assume that the attached DRA file is still active.

In addition, under the above settings, you can’t open another version of word while Dragon is working on the first. That’s why you get the error. However, you can also get the error, which can be a false positive, relative to the NatSpeak Word AddIn (dgnword_x64.dll or dgnword.dll, depending upon whether you’re using the 64-bit or 32-bit of Microsoft Office). It doesn’t mean it’s not loaded it just is slow to initialize in the second instance. It may in fact be fine, but we can’t assess that without your confirmation, version of Microsoft Office, and whether that version is 32 bit or 64-bit.

Also, when looking into that option in the Dragon Options dialog, make sure that you have not selected to run accuracy tuning by schedule or have a checkmark in the checkbox under User Experience Collection. In the latter case, that can slow your system down to a crawl no matter what you’re doing and it only benefits Nuance.

One other possibility is if your system is only set up with 4 GB of RAM. If that’s the case, you may be running into a situation where there is insufficient RAM available and Windows is swapping information and/or applications out to the Windows pagefile in order to complete the saved process. If this is the case, Windows will become extremely lethargic saving your documents will take several minutes at best.

These are the most frequent causes of this problem, but without your answer to the questions above, we can’t be sure. And, we can provide you with a more definitive answer.

Lastly, send me a copy of your Dragon log (i.e., Dragon.log(s) file(s) because in DPI 15 there may be more than one). You can email them to me at chuck@voiceteach.com. I’ll be glad to take a look at them because they will tell me almost everything that I need to know about your working system and what Dragon is saying about these processes.

Thanks again for your response. I deleted all the DRA files and reset the “save recorded dictation with document” to Never. I thought that after I did that, that all the Word documents that I’ve been working on would save faster, and they did. However, each save and autosave on those documents would re-create DRA files, despite the new settings, still taking longer than a routine Word save.

I even copied the text from one of the offending Word documents into another program and deleted the original Word document. I then shut down Word, reopened a brand-new document, and pasted the text back into the new document, but even that document is saving the DRA files upon save.


I currently run DPI 15, Knowbrainer, and voicecomputer and use them in my orthopedic practice. Ive noticed that if I load DPI with voicecomputer and then load knowbrainer, certain commands are not available in knowbrainer, eg., using the transfer command in knownbrainer to transfer text from notepad to the currently used application. I find if I load the programs in the sequence as noted above, this ‘transfer’ command does not work (cut/paste text from notepad to active application and reopen notedpad). If I load knowbrainer with dragon first, and then load voicecomputer, certain voicecomputer commands do not work but the knowbrainer commands do work (eg transfer). Any suggestions? Comments are appreciated.


While running VoiceComputer and/or KnowBrainer along with Dragon will not cause Dragon to crash and there are no negative consequences Re: any commands failing. We cannot guarantee that they will absolutely function together without issues such as the one that you bring up.

If you open Dragon with VoiceComputer and then open KnowBrainer, some KnowBrainer commands will not execute. If you open Dragon with KnowBrainer and then bring up VoiceComputer, some VoiceComputer commands will not execute. Unfortunately, there is no way that we can make them play 100% nice with one another. Part of the problem is the way that Dragon listens for commands. That is, whatever is loaded first will take precedence over the other. Therefore, if you launch KnowBrainer and Dragon, and then launch VoiceComputer, KnowBrainer commands will take precedence over KnowBrainer commands. This isn't the case with all commands, but it is related to some commands by virtue of which third-party app is launched first. On the other hand, if you launch Dragon with VoiceComputer and then open KnowBrainer, VoiceComputer commands will take precedence. Again, there is no way that we can fix this because it's the way Dragon works Re: commands.

Also, VoiceComputer is tightly integrated with Dragon. What that means is that you cannot close VoiceComputer and leave Dragon open. You must use the VoiceComputer Control Center and "shutdown speech" in order to close VoiceComputer. Because KnowBrainer is based on Dragon's Advanced Scripting, KnowBrainer can be closed down without shutting down Dragon. However, simply closing KnowBrainer and launching VoiceComputer can cause issues over which we have no control.

For example, closing KnowBrainer and launching VoiceComputer can cause VoiceComputer to fail to load. If and when that occurs, VoiceComputer may fail to launch the VoiceComputer Control Center, in which case you won't have access to the "restart Dragon" option in the Control Center. Under this condition all you can do is either manually shut down VP.exe and natspeak.exe from the task manager. Or, in some cases, you may have to restart your system.

What I recommend is the following:

1. If you want to use KnowBrainer, start KnowBrainer and have it load Dragon. You want to use VoiceComputer under this condition, use the KnowBrainer command to "CLOSE KnowBrainer and Dragon, then launch VoiceComputer.

2. If you want to use VoiceComputer, launch VoiceComputer and let it launch Dragon and when you want to switch to KnowBrainer open the VoiceComputer Control Center and select "shutdown speech", then launch KnowBrainer.

Again, anything is possible, but making KnowBrainer and VoiceComputer work together without conflict simply isn't possible on the one hand, and not something that we have the time or the resources to be able to do. Unfortunately, VoiceComputer is provided as is out-of-the-box.

VoiceComputer has thousands of commands, which are now hidden because we consider Intag technology a superior alternative than using thousands of commands.

Thus, VC probably has commands (or had commands that could be resurrected) that you could use instead of KB commands.

If you're interested, let us know some of the KB commands that are important to you.

Ron and Chucker,

Thank you for your replies. I much appreciate this.

The only command that I can think of at this point that I miss when launching Dragon with voicecomputer, that is part of KnowBrainer, is the" transfer" command where I transfer dictated text from the notepad to the other open window, as in my case, my EMR (Greenway primesuite). Once the text is transferred, the notepad reopens ready to go once again. The reason I use Notepad is that I can use all my commands (Dragon and Knowbrainer) when I dictate. If I dictate into my EMR directly (runs in Citrix), I cannot use any of these commands, as they are not recognized. Does voicecomputer have a similar command?



The command that you using and KnowBrainer is called "Transfer Back".

Although it's a simple command there are a couple of requisites and at least one caveat:

1. Focus has to be on the text window into which you want to transfer the text dictated in notepad before you even open notepad. If not, you can always "paste that", at least I believe you can, but that represents an extra step. The same applies to the Dictation Box as long as you have the settings for the Dictation Box set to "Keep transferred text in clipboard when the Dictation Box is closed".

2. Notepad is fully supported by Dragon and/or KnowBrainer. However, you're limited to plaintext. That is probably all you need in your case. Nevertheless, the limitation is that you can't use "bold", "underline", "strikethrough" etc. when working in notepad. Also, if you change the font or the font style, it will apply itself to everything (i.e., all the text) you dictate. So, as long as you're dictating plaintext and you don't care about bolding, underlining, or changing the characteristics of individual words, you're fine. The advantage of using the Dragon Dictation Box is that you have access to all Dragon's formatting commands (i.e., underlining, bolding, changing the font size and style of an individual word or phrase etc.).

3. The other KB transfer commands using DragonPad etc., are not as reliable, at least in my experimentation.Nevertheless, "transfer back" does have the advantage of reopening Notepad so that you can continue. This is not the case with the Dragon Dictation Box. You'll have to say "show Dictation Box" each time you want to continue.

4. Another caveat is that there are several other commands Re: Notepad that can get you into trouble when using Notepad. For example, there is a command for "Set Size " and "Word Wrap". These can misfire or execute without you knowing it (i.e., the consequences). If you accidentally say "set two", you can accidentally trigger the font size set to two points. If you don't know or understand what's happening, it can be extremely irritating. In this case you have to say "set twelve" or whatever font size you had or want in order to reverse the command. In the case of "Word Wrap", this command will toggle word wrap on and/or off depending upon the current status in notepad. However, in this case it occurs automatically without any notification of what happened. And, there are some other irritating issues that can be confusing.

We could locate this command for you in VoiceComputer and do so without the caveats, but we would have to test it thoroughly in order to ensure that there are no unwanted caveats.

Chucker2017-01-15 09:49:21


I have been playing around with KnowBrainer and VoiceComputer in terms of which to load first and what works if both VoiceComputer and KnowBrainer are active.

What I have found is this, and this should be noted by all users of both VoiceComputer and KnowBrainer.

First, if you launch KnowBrainer first and then attempt to launch VoiceComputer, VoiceComputer may not launch and the VoiceComputer Control Center may not display. In this case you would have to use the KnowBrainer command "close KnowBrainer and Dragon and then launch VoiceComputer. Whether or not we could correct this is unknown at this time. Even so, it would likely take a considerable amount of time to rectify the situation.

Second, if you launch VoiceComputer first and then load KnowBrainer, some of the KnowBrainer commands will be overridden by VoiceComputer. Note again that there are no negative issues per se (i.e., either one or the other will crash under normal conditions). However, VoiceComputer commands will, in some cases, override KnowBrainer commands, such as "DISPLAY numbers" will call up the InTag numbers.

Third, my testing has found that the "Transfer Back" command in KnowBrainer does work with one caveat. That is, it will transfer the content of notepad to the clipboard, but it will not transfer it directly to your text window. However, if you go to your text window and say "paste that" the text will be pasted into your text window. In other words, "Transfer Back" will not automatically transfer the text, but it is stored in the clipboard so that you can paste it into your text window.

On the other hand, there are times when it does work properly. Nevertheless, whether or not it works is intermittent and not something that I can pin down as to when it works, when it doesn't work, and why. Currently for this post I have both VoiceComputer and KnowBrainer loaded and the "Transfer Back" command for KnowBrainer seems to be working properly.

Given everything that I have posted in the previous post, it would behoove you to try this and see if it works for you. It is currently working fine for me. The caveat here seems to be the necessity to make sure that the focus is where you want in your text window before using the command "Transfer Back". Otherwise, the text in notepad will not be transferred as noted above. So, the problem is not whether the command works, it's whether it''s set up properly.
Chucker2017-01-15 15:44:05


on my system KB 2017 is interfering with VC when it’s being launched last.
The intag features won’t respond anymore.
If I first launch KB 2017 and then launch VC I detect a general slowness in my speech in general.
Any ideas?



First, the way Dragon works in terms of listening for commands is determined by the type of commands and whether or not KnowBrainer or VoiceComputer is loaded first. Therefore, if you launch KnowBrainer and then open VoiceComputer, the KnowBrainer commands will take precedence. There is nothing that we can do in such cases to make both applications play nice with one another. It's just the way Dragon works.

Second, I can't duplicate your problem with InTags not functioning when loading KnowBrainer last. Works fine for me.

Third, consider that water weighs 63 pounds per cubic foot. If you load 2 ft.³ of water into a barrel it's going to require more force to lift the barrel. To a certain extent whether or not there is enough umph in your hardware configuration, there is always the possibility of having insufficient umph to lift the barrel, so to speak. For example, I'm currently running VoiceComputer and KnowBrainer 2017 on my Core™ i7 4770K with 32 GB of DDR3 1866 MHz RAM and 8 MB of L3 cache.I don't experience any slowdown in dictation, nor do I experience any conflicts with InTags. I do notice a very tiny, and I stress TINY, impact on the execution of InTag commands, but given that the barrel ways 126 pounds, that doesn't surprise me. Nevertheless, I only load both for testing purposes. If I'm running on my system I either launch VoiceComputer, which is significantly better in many cases, or I launch KnowBrainer for looking at certain commands as well as testing their functionality. I never run both at the same time except in this instance to test your issues.

Lastly, it doesn't make sense, in the long run, for us to make VoiceComputer and KnowBrainer play nicely together. As good a relationship as I have with Lunis, it doesn't make sense for him to make both products play nicely together either. Neither of us have the time or the resources to do that.

Our responsibility is to report what we find relative to both products running together, neither of us can fix them. The primary reason for that is that they are completely different and Dragon is what it is relative to command execution.

We're not trying to tell you what to do or what not to do, what third-party applications to use or not to use. However, consider that our emphasis has to be on VoiceComputer, and we provide that support as is, out of the box.

Since I know, under the circumstances that you report, that I can't duplicate your problem, with only one very, very minor issue, which is understandable given the conditions, all we can do is report what we find.

Chucker2017-01-16 10:05:07


Hope to have a new release tomorrow that’ll have a couple of transfer commands for you to use.


my system is relatively similar to yours, I will do some testing on my girlfriends computer which has a system of about one generation before mine, see how that works. Of course it could also be an installation issue ( software conflict having something to do with other programs installed).

I don’t remember having this issue with KB2012, but anyway I’m using KB 2017 mainly to get ideas for commands that I then convert to standard Dragon commands, preferably DVC (then there is no speed loss at all).
And of course my priority is VoiceComputer :wink:

Ron, will the adjustable overlay colors be included in the new release?


One of the negative things about some of the commands in KnowBrainer is that you can't just copy and paste. For example, KnowBrainer has a nifty command called "add angle brackets to . There are two problems with this command: (1) WinRap basic in KnowBrainer uses commands that are not available in Advanced Scripting, and (2) it includes some command lines that are unnecessary, problematic and useless. I rewrote the command in Advanced Scripting such that it is a single SendKeys command. Works like a charm, executes instantly and removes all the BS lines.

Sadly, there are more of these types of commands in KnowBrainer than anyone of us would have time to rewrite. For example, in the command above KnowBrainer stores the results in the clipboard. I don't want that because it creates problems with copying and pasting to and copying from the clipboard in other areas. Lunis' command has five unnecessary and problematic command lines. I simply got rid of all that and created a simple one that works much better and much faster. I'd be more than happy to share it with anyone but it requires more testing.

New to VC/Dragon. Finding my way through InTag & will move on to dictation a bit later. Using dragon 15 w/VC, Windows 10. Issue with InTag. Often get into a cycle: I say #, dragon box says “please say that again” & InTag #s disappear. I repeat # & dragon box spells out the correct # & InTag #s return, but nothing happens. Now, sometimes on the 2nd go round the requested InTag # gets clicked and all goes well from there. Sometimes it can take 5 - 6 cycles for a click on the link, sometimes its endless. Happens across browsers. I noticed, for example, if I say 13, the InTag #s disappear on the “Thr” sound, before the “teen” gets out.

I'm editing this reply.

I thought that you were experiencing a bug I've run into occasionally but on second look, I'm not too sure, so I deleted the original post.

Please email me your VoiceComputer Log.txt located in


RonK2017-01-17 12:49:20
<div> </div><div>Bryan,</div><div> </div><div>First, I think you're talking about the Results Box, which is the small box with the yellow background that displays wherever it's positioned by default or wherever you anchor it by virtue of the Dragon Options dialog | Appearance option settings.</div><div> </div><div>Second, in the Dragon Options dialog | Commands option there are two selections (checkboxes) for:</div><div>1.  Require "Click" to select menus</div><div> </div><div>2.  Require "Click" to select buttons and other commands.</div><div> </div><div>If these two checkboxes are checked, then when every you are using InTags you have to precede the </div><div>number of the tag with "click". For example, you can't just say "13" you have to say "click 13". </div><div>Otherwise InTags won't work under that setting.</div><div> </div><div>This can happen if you're working in Mozilla Firefox. There are times when Firefox will lose its focus when you say a number (tag) and what will display is the 8 x 8 grid of InTag numbers. This can be a tricky beast and you can blame it on Firefox.</div><div> </div><div>Based on your description, this is what I suspect is the case (i.e., Firefox). I base this on the fact that you say that it sometimes can take 5 to 6 cycles. I have never seen this occur in either Google Chrome or Internet Explorer. I see this happening occasionally in Firefox, and other users of Firefox have experience this as well.</div><div> </div><div>Just take this into consideration when testing and let us know via a step-by-step description of what you're doing and what happens after each step.transfer</div>


> will the adjustable overlay colors be included in the new release?

Won’t know for another day. But, it will let you set the transparency so you’ll be able to have a completely transparent background.

> I’m using KB 2017 mainly to get ideas for commands that I then convert to standard Dragon command

Why not use AutoHotKey instead of Dragon or KB commands?

AutoHotKey is much easier to use, an add-on even lets you record macros (which you can then edit) and it’s much more powerful unless you’re a programmer.

You could use keyboard shortcuts (and thus voice commands) to activate AutoHotKey macros.

If there’s much interest in this and users start using AutoHotKey, we could integrate AutoHotKey with VC so that AutoHotKey macros would be recognized by VC without creating keyboard shortcuts.

The other thing I like about AutoHotKey is that it would mainstream voice users with other users who just want to be more productive. - You could use their macros and vice versa.