Hi,
The easiest way to do this is to scan the entire recording with the standard settings, then select the spoken section and scan with more aggressive settings. IN fact, not only is that the easiest way, it's the best way.
In general, the second scan will override the first, hence the recommended order of doing things.