There was some discussion about integrating SponsorBlock into AntennaPod (a popular open source podcast app) a while back, but it didn't go anywhere. Ajay (developer of SB) was very open to the idea, so anyone interested in developing ad skipping technology for podcasts should probably get in touch with them. I think getting a proof of concept podcast app that skips ads should not be all that difficult.
I wonder if you could train some machine learning model using the data from SponsorBlock and achieve good results on podcasts as well. That way you wouldn't be dependent on a crowdsourced online database for your offline listening. Alternatively, even creating a transscript using something like [1] and scanning for words like "sponsor", "ad" or specific company names might already be a good enough heuristic.
When I learned about tune finger printing, I was running that to identify the little motif that a podcast I listened to often, would play before and after their 'mid roll' of ads.. I thought that could be a good way to automate such things, but then they just seemed to change that tune every few months.
