The hardware specs don't tend to make that much of a difference except for the microphone. The microphone is the key and getting a true noise-canceling microphone is helpful.
Additionally, if you get a headset microphone (one you wear on your head so the microphone is always in the same place) it will also work much better.
The processor, RAM and sound card usually don't make a difference anymore (and haven't for about 8 years) for this kind of program but the microphone does.
That being said, sound cards are very very cheap. You can get a new soundcard for as low as $10 and one that will certainly work for you will retail as low as $30 and can be bought used for about $10. On ebay I am seeing them for 4.95 with about the same amount for shipping and they are new.
So I'd worry more about the mic. A cheap mic is very frustrating with voice recognition software.
Thanks Robert.
I just put in an Amazon.com order for the DNS software, a NIC card for DSL so I can quickly download the speech files, the Plantronics digital microphone and a Microsoft Office Upgrade.
In about three days I'll have it and will be back with questions if there are any installation problems.
I also wrote to Nuance and suggested they update their website's minimum system specs so there will be a lot less confusion for their prospective customers.
Thanks again for the helpful answers.
Witness! One who makes no effort to understand the situation has no basis from which to criticize the situation. -- Timberlandko (rest in peace)
Chumly wrote:Since you don't appear to believe me, you are going to find out for yourself that Dragon Naturally Speaking will work with virtually any reasonably modern PC hardware configuration, virtually any decent headset, virtually any sound card (integral to the mother board or separate it does not matter), virtually any sound system, etc.
I've used it on four different machines, two of which were 7 years old. I use the latest greatest version of Dragon Naturally Speaking and I use it almost daily.
Dragon Naturally Speaking is simply not that resource intensive an application, and it's not hardware specific beyond reasonable norms.
Dragon Naturally Speaking is simply not proprietary in its hardware requirements within the PC world.
Your automotive analogy would be the equivalent of worrying about which brand of oil to buy: Quaker Sate or Castrol.
Okay, I now have the Dragon Naturally Speaking Software and a digital microphone installed and it works great. Had no installation problems.
It leads to another set of related questions. It's a little complicated. The gist of it is simultaneous transcription through use of the same headset/mic combo, and separate files and programs so I can hear and record simultaneously using different programs and the same headset. If I had the money to do it, I would just purchase an iPod and transfer the mp3 files to it for playback with an ear bug in one ear while wearing the microphone headset and using voice recognition to transcribe it to text for closed captioning.
I am told it is possible to do this on a Mac by just re-assigning ports, but I have a PC running Windows XP and wish to do the same thing but the person advising me doesn't know how to configure the PC hardware for it. Can it be done?
Here's what I want to accomplish:
I want to do simultaneous transcriptions by playing an audio mp3 file on my PC using a footpedal for playback control and "Express Scribe," hearing it in the earpiece of my headset while using Dragon Naturally Speaking to simultaneously transcribe it in my own voice to text file using the microphone portion of the same headset (or a second headset). The headset is a digital one with both USB and dual jack plug connector options that I can plug into either the USB port on the front of my PC or into the audio jacks in the back of my PC.
What do I need to do to be able to hear audio from one file with the headset while recording my own version of that audio into another file by using Dragon Naturally Speaking and the same digital microphone/headset combo to do the listening and speaking without having to constantly pause audio, swap windows and turn the microphone on and off?
Since someone will probably ask, the reason I want to do it this way is so
I can fully train DNS to recognize only my voice rather than needing numerous User profiles for the many sources of recordings I'll be transcribing, especially those with multiple voice interviews in the audio playback.
Thanks for the help. If this is a mystery to you, can you suggest some of the more obscure sources to check into for help? I've done a Google search and not found much.
I don't get what the problem is. Have you tried it? If so, what doesn't work or should work better?
E.g. if you play the mp3 files on Windows Media Player can you hear it on your headphones and use DNS?
If so, you could make it minimize to tray so you have the controls there without needing to switch windows. But if it doesn't work what's happening?
The audio input (microphone) and audio output (speakers/headphones) operate independently.
So long as your foot pedal will stop and start the playback, there shouldn't be any problem.
I think I have it figured out. Drew Dad is on the right path. It didn't work while I was using the USB port to connect the headset to the PC. As soon as I moved the headset to the jacks in the back, the earphones, microphone and USB footpedal are all finally working simultaneously. I'm using the footpedal for playback controls instead of using the mouse for it. Now all I do is click the mouse in the recording/transcribing window when I want it to be on top to catch the cursor while it transcribes ,and use the footpedal to rewind and forward for the next phrase to translate into the recorder.
What was taking me nearly 10 hours to transcribe manually, is now down to 3 hours thanks to voice recognition and simultaneous transcription. What a difference!
Now I can get it into production.
Thanks everyone! I don't understand why it didn't work when the headset was plugged into the USB port, but at this point, I don't care. It's all working another way! It must have something to do with the independent input/output functions being split with two line jacks vs. the one USB mixing it with one connector.
