any tips for smoothing result from stitching together multiple interviews?

I've found that if you have a long audio file with consistent audio--for example, from audiobook narration--you tend to get pretty good results from as little as around 45 minutes of audio. But if you're stitching together many interviews, even if they're quite good quality (from professional radio/podcast interviews), they're maybe just different enough that the final result sounds too mechanical.

Like, I've got one I'm working on, and it's 90% of the way there, but I just can't get it that last 10%. Mostly it will be fine, but then it will got a little buggy on certain sounds, etc. But this is building off of nearly 3 hours of audio that I fully stand behind.

So does anybody have any tips for smoothing this out?