I have 2 files which contain stimuli that I am using for a research study. The stimuli consist of masking noise with a 1K target tone. One presentation of the masking noise has a 400HZ wide spectral notch and one does not. I need to be sure that those are the stimuli that are being played because I am not getting the responses that I expected. I was told to do an FFT analysis on my output. In audicity I did a “spectrum plot” but I am not sure what I am looking at. For example, on the graph that shows db level vs frequency, the dB levels are negative. Negative relative to what? I’m really not sure what I’m looking at. If anyone can help me it would be greatly appreciated. Thanks.
In digital audio, “zero” is the maximum possible volume. That’s where the conventional digital sound system runs out of numbers and just stops increasing. Audacity uses special “32-bit floating” internally so it can increase beyond zero and then recover later. This is needed because some of the tools cause a surprising increase in volume and Audacity must be prevented from destroying a show if that happens.
Most other sound systems like Music CDs and Digital TV audio have the zero restriction because of size and transmission constraints. On the Audacity sound meters this is all the way up toward the right and the red lights come on. On the blue waves, it’s represented by 1.0 or 100%, either up or down.
Analyze > Plot Spectrum can be used to tell where the energy is in your selection. The detail and information gets better as you pull the window larger. The SIZE setting is of particular importance because the accuracy of the plot goes up with size. However, it’s possible to get so much detail in the plot that you can’t read any of it. Juggle as needed.
In the illustration, I have created a noise (all frequency) “show” and then applied a notch at 1000Hz with a quality (width) of 10.
Koz
Thank you. I will try it and see how it goes. I appreciate your help.