I have tried many microphones over the years to fit my particular needs. My latest acquisition is the Plantronics CS-50, and I am having great success with it. It's range is incredible, and so far, its noise canceling capabilities must be pretty good as the television set does not cause Windows Speech Recognition to shut down randomly like it has always done before. Anyway, this is only secondary to the reason I'm posting this; for the same number of years that I have been looking for the perfect microphone, I have been trying to figure out the proper recording device settings for each microphone I've tried.

I now know that this Plantronics microphone is my best option, but nowhere on the Internet can I find any real information on the best settings for recording devices where speech recognition is concerned. For instance, this microphone has multiple settings for sample rate and bit depth, ranging from one channel, 16 bit, 8000 hz (telephone quality), to two channels, 16 bit, 48000 hz. What I've never understood, and I've never been able to find out, is what is the best choice for speech recognition microphones? Also, under the advanced tab in microphone properties, there is another option called Exclusive Mode (allow applications to take exclusive control of this device & give exclusive mode applications priority). When using a microphone for speech recognition, should one turn these options on, or turn them off?

If anyone has an answer to either of these questions, I would certainly appreciate hearing it. :-)

MacktheKnife wrote: I used to use Dragon several years ago. Of the reasons for stopping was the headset that came with Dragon broke and accuracy was strangely good at home but not at my office. Both environments were quiet. At home the computer was an old IBM Thinkpad with Windows XP. At the office, the comuter was a newer Dell tower with Vista. I'm reading things about a USB sound pod but not sure that would make a difference? I read the getting started with windows speech recognition article and would like to try it. Any suggestions on a good microphone that is not too expensive yet will give few errors in my quiet offices?

1. A good microphone that not only send the purest audio, but also rejects backgroun noise.
2. A good sound card that takes your analog speech waves and converts these to digital bits and bytes a computer understands. A USB sound pod is simply a sound card that sits outside the computer. Audio from the microphone is passed directly into the computer CPU, bypassing electronic noise within the computer.
3. The human factor. This includes:

• Good enunciation. Each word should be clearly spoken, separate from other words.
• Speak in phrases. Because speech recognition software uses context of surrounding words as in, "They're going to park their car over there," speaking in phrases helps the software figure out which homophone to use.
• Understanding how to use the program and make corrections. Corrections improve your user training profile. Most speech recognition software has tutorials built in. It is well worth the time to go through these as it will save you much time later on.

The 2 headsets under $100.00 to look at are the VXI Xpressway with USB and the Andrea ANC-700/USB

Marty Markoe, eMicrophones, Inc.

I used to use Dragon several years ago. Ofthe reasons for stopping was the headset that came with Dragon broke andaccuracy was strangely good at home but not at my office. Both environmentswere quiet. At home the computer was an old IBM Thinkpad with Windows XP. Atthe office, the comuter was a newer Dell tower with Vista.

I'mreading things about a USB sound pod but not sure that would make a difference?I read the getting started with windows speech recognition article and wouldlike to try it. Any suggestions on a good microphone that is not too expensiveyet will give few errors in my quiet offices?