I have an audio file of a person answering several questions. Between each answer, there are seven seconds of music before the person starts to speak again. The audio is originally from a video and during this time, the asked questions would be displayed. For my use case however, I only need the audio of the person, without the separating music. So e.g instead of one file with "person speaking-music-person speaking again," I want two files "person speaking" and "person speaking again".
As the separating music is always exactly the same, I thought it might be possible to detect it automatically and split the audio before and after. We are talking about several hours of audio, so cutting the file by hand is not really a preferred option here.
How would I approach this when trying to program something myself? This whole field of audio editing is relatively new to me, so I am not really sure where to start.
One idea I had was to convert the audio (it is mp3) to a base64 string and to search for the corresponding string of the separating music but I assume that, the way mp3 files are designed, this is not really as easy as it sounds right?