Tapesearch is a search engine to quickly search podcast transcripts. Find what was said in a podcast and go straight to that point in a recording wihout having to listen to hours of audio.
Currently the most popular podcasts based on the number of iTunes ratings are included (not the rating itself). The long term aim is to include as many podcasts as possible. If you think a podcast should be included please get in touch using the chat box (priority will be given to requests from subscribers).
Generally the AI is very good at transcribing, but you might find small errors with people's names, fast speech and when people are talking over music. Long periods of silence and/or music can also introduce errors in the timestamps which can cumulate over long recordings. If you see a particularly bad transcript, let us know and we can try a different model.
Some podcast providers like to change their adverts too. So if the advert was different to the one at the time of transcription, you may notice a difference. If the advert was different in length, then there might be an error of a few seconds in the timestamps
TXT, VTT and SRT with only the latter two containing timestamps. These are all ASCII files and can be opened with a text editor, e.g. Notepad