Speech Recognition: Software and Microphone

Discussion in 'Parallels Desktop for Mac' started by ijames@mac.com, Sep 21, 2006.

  1. ijames@mac.com

    ijames@mac.com Bit poster

    Messages:
    2
    Hello There Everyone,

    I have a client who is looking to use his new installation of Parallels and Windows to run the latest and greatest speech recognition software. I have researched a bit but 10,000 heads are better than 1. Consequently, I have a few questions. Some questions may be outside the realm of the parallels team. Some questions should be within your sphere of concern. Any help that you can give will be most appreciated.

    1. What is the best speech recognition software that works in the Mac parallels environment with Windows XP?

    2. What is the best microphone designed for dictation that is functional with the current version of Parallels? (There is mention that isosynchronous mics do not function in parallels. Does this effect me?)

    Customer Configuration
    Mac Pro Quad Core 3.0GHz (Temporarily reduced to dual core to address Parallels kernel panics)
    2GB's of RAM (we have more on hand)
    Mac OS 10.4.7
    Windows XP, Office XP


    Thank you very much for your time,
    James L. McMahan Jr.
     
  2. billmccl

    billmccl Junior Member

    Messages:
    12
    Speech Recognition in Parallels

    I use speech recognition extensively, for hours a day.

    I own both Ilisten for OSX, and Dragon 8.1, soon going to 9.0 for Windows XP. I use a MacBook Pro at the office, and a Mac Pro at home. I have tried to run Dragon in Parallels, but have not been successful. The problem is the microphone in Parallels, not the software. I use the Philips SpeechMike Pro for Dragon, and the Plantronics .85 for Ilisten in OSX.

    I routinely run Dragon 8.1 in Boot Camp using Windows XP with the Philips mike, indeed, it is now my favorite platform since my MacBook Pro and Mac Pro are the fastest Windows machines at my disposal. Dragon is the only residual application I use on a Windows machine, and I would love to get reliable industrial strength voice recognition over to the Mac platform. I would offer the following observatioins:

    1. My preference for Dragon over Ilisten is not based on the software, but on the microphone. I use voice recognition for hours a day as a busy internal medicine doctor, and I just can't use the boom microphone/headset required for Ilisten for extended periods of time. I find the workflow with Dragon a little superior to Ilisten, but accuracy seems quite good with both. The Philips mike is not supported on the Mac in OSX, something to do with the USB port configuration, I think. I would likely switch to Ilisten on OSX if MacSpeech came up with a hand mike as good as Phiips, since all of my other applications run in OSX either natively or via the OSX citrix client. This includes the GE Centricity Office EMR and the Eclipsys Sunrise Clinical Manager hospital information system and our organization's Groupwise email and PIM. From a clinical perspective, we are totally paperless with no paper charts in either our office or in the hospital. So the speech recognition software is at the core all of our clinical activities.

    2. I had no diffuculty installing the Philips software in Parallels, but the function keys that are required to support the Phiips mike do not work. To be more specific, the Philips microphone generates function keys to transmit commands to the software, and Parallels just can't get these comands, no matter what I do. I left a message here on the forum, but never got a satisfoctory response that worked to capture the function keys generated by the Philips microphone control buttons in Windows XP in parallels, or to get the microphone programed for other keys. I suspect this can be made to work, but I gave up since Boot Camp supports it so well. But I would prefer Paralles over Boot Camp if it could be made to work.

    3. I have not tried to use Ilisten in Citrix on OSX, and this might be problematic, I just do not know. I have talked to the tech support people at MacSpeech, and they were not very helpful, nor had they apparently even tried to do this. I think they are missing a huge opportunity, and still have that smug Mac only arrogance, not recognizing that there are a huge number of people like myself who are compelled for business needs to remain partially in the Windows world, but desiring to maximize the OSX esperience. Obviously, Paralles understands this.

    3. I would love to use MacSpeech's product since I am a confirmed Mac user, but I just cannot get over the microphone issues. Macspeech is kind of the Avis against Hertz vendor, and they are almost there. Their software is fine. But they have not tackled the microphone issues, leaving it to the hardware vendors, who have very little economic motivation to solve the problems for the modest number of Mac business transcription users. I think that the boom mic is adequate for the casual user, probably for the student user, but the workflow enhancements avialable with function buttons on a hand microphone will in my judgment lead the serious transcriber away from MacSpeech and Ilsten until they solve this, no matter how good their software.

    4. Getting the Philips microphone with Dragon to work on Windows XP in Parallels, complete with microphone function keys would be a great accomplishment. I hope that someone has done it or will do it soon, and that they share that knowledge with the rest of us.

    Hope that this has been helpful.....
     
  3. ijames@mac.com

    ijames@mac.com Bit poster

    Messages:
    2
    Very Helpful

    Hi Bill!

    Thank you very much for your thorough response. It is quite helpful. I do have a couple questions. What functions are provided by the function keys? Is the mic workable for straight dictation without those function keys or is the mic TOTALLY non-functional in parallels? When testing, have you had any issues with speed of dictation in parallels?

    Thanks Again,
    James
     
  4. billmccl

    billmccl Junior Member

    Messages:
    12
    I cannot answer for the partial audio functionality of the Philips mike, have never gotten past the configuration of the software buttons to give it a try. And as I said, without the function keys, the mike is no better than many other boom/headset mikes out there.

    Imagine 6-8 hours a day in transcription, with many interruptions. Think of a radiiologist, for instance. He may need to trasncribe 30 xray reports an hour. His living is dependent on it. We refer to workflow vs technology frequently in working with doctors. The Workflow usually trumps the technology. The doctors talk about the "click count" of a given application, and actually measure it. The Philips mike has a mouse built into the handle, complete with rt and lft button and a track ball. Great for navigating an application or a document. The buttons (a total of 10 including the mouse buttons), can be configured to the user's taste, but our configuration for our doctors in Dragon includes on/off of the mic input (think about transcribing, then being briefly interrupted for a question etc. You don't want the casual conversation to be picked up), and various "go to" commands in the text of your transcription for placing of the cursor in editing, ie go to end of line, end of paragraph, end of document, beginning of document, beginnning of line etc etc. The language of communication of the speech mike command buttons is the function keys. The speech mike generates a function key command that directs the application to some action, like move the cursor. F11 directs the mike to go off, F9 for the mike to go on etc. These function key commands are generated by the microphone, not the keyboard. The trasncriber never has to touch the mouse, the keyboard, or the application, everything is done in the microphone which he holds in the hand.

    If you look at some of the stand alone transcription systems like Lanier etc, you will see a similar functionality built into their microphones.

    Again, my expectation, (and hope!) is that someone out there in Parallels land can sort this out and get it to work. I suspect there will be some issues with the USB port also, but since it works flawlessly in Boot Camp, I can only assume it is a software issue with the OSX implementation of the USB standard, as well as the frequently flawed Parallels use of the USB port.

    I have not tested accuracy in Dragon in Parallels, have not had a reason to do it. As I said, the microphone is everything for the medical transcriber. Dragon is a cpu hungry app, though. Our Windows machines are typcally spec'd out with a minimum 3ghz CPU and 2 gb of RAM. So there may be issues with performance in Parallels, I just haven't got there to find out.

    Be curious of the results if you put some energy into further testing of the Phililps mike, would love to know if you ever get it to work satisfactorily. Again, my view is that the Mac world is a market in waiting for an industrial strength microphone/voice recognition application, even if it is running through parallels. But until these issues are addressed, our friends in Seattle will rule this space.

    Bill
     
  5. bob stern

    bob stern Bit poster

    Messages:
    1
  6. ithacalaw

    ithacalaw Junior Member

    Messages:
    17
    I am also a physician (internal medicine/endocrinology). I am using ComChart Electronic Medical Records system (http://www.comchart.com/ ) which as Filemaker based program written by a physician - Hayward Zwerling. I have purchased a Mac Pro/Dragon NS 9/Philips Speech Mike Pro/Parallels and XP pro. I have managed getting the DNS9/Philips Speech Mike working fine on a Windows box. I am planning the next step this week. I feel that jumping in and seeing what happens is the best way to go. I could use boot camp and just use the Mac Pro as a Windows box, but that would be defeating the point. Other approaches include using CodeWeavers CrossOver - tho' this doesn't support DNS9 yet I suggest you go to this URL and Vote for them to support it (it is currently ranked number 75 in the list of Window's applications that users have voted on for them to make compatible).
    http://www.codeweavers.com/compatibility/browse/name/?app_id=2092;vote=1
    However, I would like Parallels to win in this market.
    Let me know which electronic medical record system you are using and how you incorporate text from voice recognition into your records - do you cut and paste it in? do you directly dictate it in? do you use a mixture of templates and dictation?
     
    Last edited: Oct 21, 2006
  7. sgruttadauria

    sgruttadauria Bit poster

    Messages:
    3
    Hello,
    I am a Doctor of Audiology and have been using Dragon 9.0 medical within the windows platform. I have installed Parallels and Dragon without a problem. I am using a Platonics USB mic. It shows up under "devices" but when I go to check it, it says that it is busy. I can't do my dictaton at all. Any help would be appreciated. Would I have the same issues if I used Bootcamp?
    Thanks
     
  8. unused_user_name

    unused_user_name Pro

    Messages:
    495
    Don't attach the Mic to Windows at all. Don't tell Windows it is a USB Mic.

    Select your microphone as the input in MacOS, the VM will automatically see it as "Mic In" when you enable the VM's sound card.
     
  9. jonnib

    jonnib Bit poster

    Messages:
    1
    I am trying to use a speech recognition product for reading improvement that I am developing - its currently WinXP only and its killing me not be able to use it in parellels

    I have my USB headset selected as the line in and out on my mac OS but only see Intel integrated audio in my Sounds and Devices properties on the VM pc. I have sound enabled checked in the properties.

    Any help would be a godsend.

    cheers




    MacBook 2.0Ghz, 2gb RAM, 80gb Disk
     
  10. unused_user_name

    unused_user_name Pro

    Messages:
    495
    Alright,

    Here we go again....

    0) Make sure your Parallels instillation has the VM tools installed, and a sound card that is enabled.
    1) Set the USB device as the input in Mac OS. Set the input volume in MacOS.
    2) Launch Parallels
    3) In the VM, select the "Mic in" as the input. The VM will not know anything about a USB headset. This is normal.
    4) Use the microphone.
     
  11. eseda

    eseda Bit poster

    Messages:
    4
    Dictaphone Powerscribe

    Has anyone had any success in getting a Dictaphone Powerscribe Microphone to work in Parallels?

    This is a USB mic also, but the Dictaphone software (Boomerang) appears to require USB recognition for the mic buttons to work. When I try to add this USB device Parallels gives an error, USB device is in use by other application.....etc. I do not see any Mac OS app using the USB mic and have tried setting the Mac audio to USB mic, to internal and to Line In.

    I am using Bootcamp and XP all the time and it works, but I would like to make use of the Mac OS with Parallels. I tried Crossover Office for the USB mic, but it does not see the mic. I dictate all day as a radiologist and I have to use this Dictaphone system for our clinic transcription.

    Any suggestions?
     
  12. sgruttadauria

    sgruttadauria Bit poster

    Messages:
    3
    Hello, thanks for the reply, but I am still having trouble and feel pretty stupid.
    1. How do I make sure that the sound card is enabledd? (I have installed the VM tools)
    2. Where do I select "Mic in" in the VM. I just have "integrated audio" listed.

    Thanks. Your help is really appreciated.
     
  13. unused_user_name

    unused_user_name Pro

    Messages:
    495
    1) In the VM prefrences pane, you have to make sure that the VM has a sound card installed in it. Look at the list of hardware (where it lists your hard drives and network interfaces, make sure there is a sound card listed. If there isn't one add it)

    2) In the recording side of the windows volume control. (Open volume control. Goto prefrences, select recording, press ok. Make sure Mic In is selected)

    I don't have this hardware so I can't try it, but you could try upgrading to 3036 with their new USB stuff. You can also manually unload the Mac driver for USB audio to make the "in use" message go away.

    To do this make use of the kextunload command. Search the forums here for it if you do not know how to use it.
     
  14. eseda

    eseda Bit poster

    Messages:
    4
    What is 3036 upgrade?

    I have the latest version of Parallels, is this the 3036 upgrade? Or does this refer to an OS10 component?

    I tried the kextunload method, but it does not work.

    When I add the USB microphone I get one new line;

    com.apple.driver.AppleUSBAudio

    I can delete the line using kextunload, however Parallels still detects the USB mic as in use. (See my posting in the kextunload thread for details)

    Is there a different method to get OS10 to let go of the USB mic, or is there another line that I should disable in kextstat?
     
  15. unused_user_name

    unused_user_name Pro

    Messages:
    495
    3036 is the latest beta. There is a link to it in the stickey at the top of the forums.

    As to the driver issue, my guess would be some other driver must be using the mic as well. I'm not sure here.

    I tried this initially and got the mic to connect to Parallels, but in my case the microphone did not work as the USB converter I am using requires USB 2.0 and would randomly disconnect in Parallels.

    I use my mic through the Mac sound system.
     
    Last edited: Dec 11, 2006
  16. eseda

    eseda Bit poster

    Messages:
    4
    Dictaphone Powerscribe

    Thanks for the help.

    I tried the 3036 Parallels Beta last night, same USB device is in use error.

    The Dictaphone Mic must be a hybrid USB device as explained in a prior posting regading an other USB device. This makes sense as it is both an audio device(mic and speaker) as well as a mouse/keyboard USB device for it's control buttons. The Kextstat/Kextunload method only detects the audio portion and deleting the mouse/keyboard entries is not an option as I have to use a USB mouse.

    So I am stuck in Bootcamp for now, until Parallels catches up and adds adequate USB support. This is the only application that forces me to boot into Windows.
     
  17. shadowreader

    shadowreader Junior Member

    Messages:
    12
    Any success on this? I am trying to use Powerscribe under Parallels also and usually use a Philips Speechmike Pro 6274 with Powerscribe(not supported but works fine). However, when I plug the Powerscribe mic in it crashes the Parallel session. Any ideas?
     
  18. jip

    jip Bit poster

    Messages:
    3
    This was supposed to be a quotattion from an earlier post:

    1. My preference for Dragon over Ilisten is not based on the software, but on the microphone. I use voice recognition for hours a day as a busy internal medicine doctor, and I just can't use the boom microphone/headset required for Ilisten for extended periods of time. I find the workflow with Dragon a little superior to Ilisten, but accuracy seems quite good with both.

    Now my question:

    I have a question about your observation about the accuracy of the two programs. I've been hearing both views: from the iListen people and adherents I get the view that their voice software is as good as Dragon's, but their learning curve is steeper; it is capable of equal accuracy over time; from others I hear the opposite, including from some who are impartial. Now you sound impartial, and impartial to both software apps. Did you really get equal results (accuracy) from both? Or did you not get far enough with iListen to know? (I'm trying to decide which way to go.)

    Thanks.
     
  19. hacky

    hacky Junior Member

    Messages:
    10
    I use Dragon 9 with Parallels. I think it is far superior to iListen in terms of accuracy and usability.
     
  20. jip

    jip Bit poster

    Messages:
    3
    Thanks. Do you experience any time-lags? (beyond the ordinary, that is: a) in the Dragon software itself; b) in switching from Mac to PC to Mac) Which version of Dragon are you using? (I was thinking of burying the Preferred version) And which mic? (I was assuming I would use whichever came with Preferred, and following the tips I've read on these forums). Finally, would a macmini duo core 1.6 GHz be powerful enough to drive all this, or do I need more muscle?

    A few more issues: I use ancient Greek a lot, often embedded in my sentences; also EndNote for bibliography. The Greek works best in Mac (it is a keyboard layout mapping + unicode fontsets); I don't want to buy a new endnote program again, and I don't want to learn a new keyboard mapping. I gather the way this would work is that you (I would) put my files in a shared folder, then grab them from the other (here, Mac) side, and do what I want on that side, then grab them from the other (Windows) side, and so on, back and forth. Does this mean you have to close out the file from (say) Word on either side to 'release' it to the other operating system? (That would be clunky if all I wanted to do was to add a footnote or a greek word.) On the other hand, my RSI is telling me to do something drastic. If necessary, get a hold of Endnote for windows, find a greek for windows equivalent. Hey, wait: I might as well buy a ... PC??:eek:

    But seriously, I'm just trying to envision how this works in practice.

    thanks.
     
    Last edited: Feb 21, 2007

Share This Page