ImLeigh

Speech to text has come a long way but your desire to have DNS transcribe from any voice is not what the software promises. I have a totally different idea of how DNS works! I am wanting to purchase it to augment my transcription as "once trained" it boasts 99% accuracy with recognition/transcription speeds of "up to 160 wpm" I don't type overly fast (65 wpm) and see the potential of being able to dictate faster than I type 99% seems really good BUT that means there is an error every 100 words -- that translates to one to six errors per typed page (depending on line spacing font size etc) so I expect some of the time saved will be used to proof documents carefully before they go out. -- DNS needs to be "trained" to your voice to transcribe what you dictate into a word processor or email. I envision listening to the file I am transcribing bylistening through earphones to the expressscribe file and simultaneously dictating what I am listening to into a microphone attached to my computer -- It is my dictation that should be converted to text. I use Microsoft Word and Word Perfect -- I don't use the ES notepad but could see where ES would get a bit confused re: which audio to convert to text. Of course there is always the possibility that you are trying to transcibe your own speech (& Dragon has been trained to your speech) My understanding is that you should be able to use digital recordings of your own voice to generate text from speech. Perhaps it is this that you are being unsuccessful in using Express Scribe -- You might try to find out how to play that file from the Dragon Naturally Speaking people -- it took some digging but they have a wealth of information on their website.