Audacity Forum

Greetings - I'm new to Audacity. I'm a professional transcriptionist, meaning I transcribe audio of all types into text for commercial use. Unfortunately, a large number of my clients don't understand that taping their very important interview in the middle of a busy Starbucks is a bad thing for the transcriptionist. I had a tip that Audacity might be able to help me remove/limit all of the background noise (other conversations, laughing, dishes clinking, et cetera), while retaining the integrity/clarity of the foreground voices I need to transcribe. Once processed, I would then like to upload the improved audio file into my transcription software (Express Scribe - NCH software) for transcription. I've looked through the tutorials, and the forum, and I don't see anything to indicate how this might be done. If anyone could help me out, it would be greatly appreciated.

Incidentally, I'm using the 1.3.x (Beta) version of Audacity on a Windows 7 machine.

Thanks!

Separating multiple voices from a single mono recording is virtually impossible.
What you are able to do (quite easily) is to remove sounds that are lower or higher in frequency that the voice that you are listening to. Note, this will not remove other voices because the frequency ranges of voices are all pretty much the same, but you can remove other noises such as rumbling, or high pitched noises, and doing so may help you to concentrate on the voice.

Voices are mostly in the range of about 300 Hz to 7000 Hz (7 kHz).
You can use the "Equalization" effect to cut out sounds that are outside of this range.
See here for information about the Equalization effect: http://manual.audacityteam.org/man/Equalization

The required "curve" will look something like this: