Need help with listening test evaluation - semantic differential


I am an absolute novice and hardly know anything about statistical evaluations, which is why I urgently need your help. I do not know which statistical methods are suitable for my evaluation. However, I have the programs Excel and SPSS at my disposal.

I have carried out an online listening test, in which several listening examples of a total of 4 different podcast versions were played to the participating test persons. Per podcast version 4 listening examples each. So a total of 16 audio samples. The test persons had to judge the audio samples on a polarity scale according to predefined attributes. These are all attributes that describe the sound experience (dull/clear, far away/close, unrealistic/realistic etc.). Except for one attribute that describes the overall impression with the adjective pairs "I don't like/ I like". Thus, I collected data indicating how the audio samples of the podcast versions were rated with respect to the attributes in a 7-level semantic differential. (-3, -2, -1, 0, 1, 2, 3)

Now I would like to compare these ratings (in terms of the respective attributes) of the individual podcast versions. Now I was thinking, this is probably best done with a mean comparison? Or a correlation analysis?
So e.g. I have the adjective pair unrealistic - realistic - that was answered now to the audio examples of the 4 Podcast variants in each case and I would like to know now whether the evaluation differs with the four Podcast variants significantly. --> Which podcast was judged more realistic by the test persons regarding the sound experience? Which one achieved a better overall impression with regard to the sound experience? etc.?

And then secondly I would like to determine what influence the individual attributes (dull/clear, unrealistic/realistic etc.) have on the attribute for the overall impression (I don't like/ I like). So that in the end I can say that, for example, if a podcast was judged to be very realistic, that this attribute also contributed greatly to the overall impression being high. Or just not. --> Is it important that the sound experience of a podcast is perceived as realistic, so that the overall impression is good? These kinds of questions.

I'm just getting desperate because I'm new to the field and there are so many different statistical methods out there. Would any of you have an idea which statistical methods are useful for the two determinations? I would be very grateful!

With kind regards