I have an interview recording where the levels were too low. I’ve tried using Compressor, Amplification, and Normalize to increase the levels of the recording, but the handful of peaks in the recording are hindering the more uniform amplification I could be getting from the compressor effect. You can see this in the attached image. How do I reduce the impact of those peaks so that the recording can be compressed in a more uniform manner, leading to a better usage of the recording’s dynamic range? If I select a section of the recording that is already fairly uniform and use the compressor on that, the result is that the compressed section uses a greater extent of the dynamic range (it’s louder). You can see this in the second attached image.
The problematic peaks are just normal parts of the interviewee’s speech. Thanks!
You should be able to adjust the compressor to help that. Have you tried the “Compress Based On Peaks” setting?
There’s also “Limiter” and “Hard Limiter,” but I’m not sure which ones come with Audacity. Some of those tools are home baked.
If you want to commit to it, you can try Chris’s Compressor. Chris can simulate radio station sound compression, so no matter what you do, you don’t have to reach over and change the radio volume in your car. It does have a personality, so if you like it, you may have to do them all that way. I change the first value, Compression from 0.5 to 0.77 to get a louder, tighter sound.