You’ve just got you new recorder delivered, tested it for a first recording and just discovered that on your finished recording there’s a hum or a hiss.
How do you fix it? This quick tutorial will walk you how to import your audio into Audacity, split the stereo track and remove the noise.
Start by downloading Audacity for your operating system (Mac, Linux or Windows).
Import your Audio, sample background noise
After installing Audacity, open the application, and import your sound using the File > Import menu item.
Next, take a sample of the background noise to be filtered out. With the tracks imported, select an area of the track that contains only the background noise. Then choose Effect > Noise Reduction from the menu, and press the Get Noise Profile button.
Filter the Noise
Next, select the area of the track you want to filter the noise from. Do this either by selecting with the mouse, or Ctrl + a to select the entire track. Finally, open the Effect > Noise Reduction dialog again, and click OK to apply the filter.
Additionally, play around with the settings until your tracks sound better.
If you don’t like the results, press CTRL-Z (Windows / Linux), Command+Z (MAC) to undo, change some settings, and try it again. The default noise reduction level is −24 dB, which means frequencies identified as noise are attenuated by −24 dB. If this removes too much of the recording, set a higher decibel value, such as −20 dB, and tryit again. More than likely you won’t be able to erase the noise completely because doing so may erase things you want to keep, but you can lower it to where it’s not so noticeable.The Frequency Smoothing slider is more surgical at smaller values and affects a wider range of frequencies as you move the slider right to set larger values. Try smaller values first, because they’ll affect less.The Sensitivity time slider determines how fast the Noise Removal effect responds to change in the audio signal. For noise that is fairly steady, use a larger value. If the noise fluctuates rapidly, then use a smaller value for a faster response time.