You may find it very useful to have a book containing many examples of spectra. Although old, I had continued to use Nakanishi's book as it discussed the absorptions and included many examples you could refer to. I think if you looked at examples of different types of carbonyl groups, you may be able to identify how these may differ.
IR spectra are usually scaled to 100%, so peaks are relative. A strong carbonyl group can make other peaks smaller or if they are more prevalent. If there more hydrogens in one sample than another, for example, they may appear much larger. If a sample doesn't have a carbonyl group (which often scales to 100%), the CH's may appear much larger compared to the next most intense peak.