I just got off the phone with John at MacSpeech who tells me that MacSpeech Dictate is known to be incompatible with Parallels Desktop for the Mac. (My symptoms were crazy internal capitalization within words and generally degraded recognition quality). Apparently there are no plans to make it work since "Parallels isn't really a Mac application" and in any event it's not supported. I had done a bunch of research before buying the product and had seen a number of reviews indicating that the software works with Parallels. I didn't see anywhere (including in the Dictate user manual) that indicated it doesn't. I just want to make sure the word gets out there that MacSpeech Dictate is NOT compatible with Parallels Desktop. Hopefully I'll be able to get a refund for the software from the vendor where I purchased it -- John at MacSpeech wasn't offering any recourse on their end.
Did you try it without Parallels, friend of mine has this, and he also cannot get normal recognition quality, even he has no Parallels installed on his MB
Without Parallels it seemed to work OK. I had to fiddle with the microphone a fair amount to get acceptable sound quality, but once I did it seemed to work fine outside of Parallels.
Thanks for this report! I don't follow though: are you suggesting that MacSpeech Dictate won't function normally on a Mac that has merely Parallels Desktop 3 *installed*? Or that they can't both be *running* simultaneously? Finally, what do you mean by "once I did it seemed to work fine outside of Parallels"? Why were you running it "inside" Parallels to begin with?
I was trying to use Dictate to dictate text into applications running within Parallels (a number of reviews suggested that this worked fine). I don't think there's any problem having both programs installed. The issue is trying to use Dictate as a method of inputting text into an application running within Windows within Parallels. In terms of "inside" vs. "outside" Parallels, in all cases Dictate is running in OS X. I meant when I targeted an application outside Parallels as the recipient of text input it worked just fine. The problems arose when I brought Parallels to the front so it was the target application for text input.
Dictate will not work with Parallels applications. Ok, I got the situation, I suppose high load is preenting Dictate to recognize words correctly, as it is CPU hungry application Try to perform everything to minimize CPU load for Parallels first thing put VMs to external drives, other steps bellow: 1. Set memory for Virtual Machine to less than 1 GB 2. Perform Mac and Windows side disk defragmentation http://www.xvsxp.com/system/system_tools_defrag.php 3. Set Adjust memory limit from Auto to Manual at Parallels Desktop -Preferences-Memory, and set limit a bit more higher than VM memory 4. Disable Themes in Windows XP serives.msc stop Themes service 5. If you are running in Coherence, try to switch to Single Window, or lower Mac resolution 6. Some antivirus software can load CPU when Heuristic Prediction is set, or Tamper protection (Symantec, MacAffey), adjust settings to risk you are supposed, high security settings is not always good idea, as it can also affect software running 7. Run msconfig , and disable unneeded startup items See http://www.netsquirrel.com/msconfig/msconfig_xp.html 8. Make sure that no VNC,or remote desktop connection is established to Mac, this can slowdown Parallels Desktop 9. Optimize computer experience for best performance
Ahhhhhh, well there you have it! Your post made it appear that the two programs couldn't coexist at all. If you want to enter spoken text into the PC environment, you'll of course need to use a PC speech recognition program such as Nuance Dragon Naturally Speaking, and have it interact directly with your USB microphone. Dictate should be turned off so it doesn't interfere.
Is Parallels fast enough to run Dragon Naturally Speaking for speech recognition? I wouldn't want to buy another program and have it fail to work because the VM is too slow. I'm running it on a 2.66 GHz (I think) 24" iMac. Thanks.
VM slowness depends from how much resources you set to it, and from Windows settings, Dragon Naturally Speaking is confirmed to work properly with internal audio devices
By "internal audio devices" you mean an audio input device controlled directly by the Windows OS, right? By calling them "internal audio devices," some might think you mean external microphones won't work -- when in fact speech recognition will work best *only* with a noise-canceling external digital microphone.