audfprint

Cool. I was playing with audio landmarks[1] last week to remove ads from podcast before I listen to them. I wonder if something like spleeter could be used together with landmarking as an elegant way to remove copyrighted music and retain speech for videos to be uploaded to Youtube. [1]https://github.com/dpwe/audfprint

My understanding is you're looking for landmarks in the audio and then matching/aligning on those. Shenidam is open source software which does the whole process, site offline, GitHub still up: https://web.archive.org/web/20160309034204/http://shenidam.o... and https://github.com/stendardo/shenidam A node stream module: https://github.com/adblockradio/stream-audio-fingerprint Python: https://github.com/dpwe/audfprint A blog post explaining fingerprinting: https://willdrevo.com/fingerprinting-and-audio-recognition-w... Commercial software: https://www.redgiant.com/products/shooter-pluraleyes/

It's all about the FREQUENCIES! Actually though, this system can be used to find copies of any sound -- sound bytes, commercials, music, whatever. We just happen to be using it for politics because we apparently hate fun. We're using the python port of audfprint though: https://github.com/dpwe/audfprint And yes -- we have a big bucket of known ads, and a big bucket of archived programming, and we're comparing new content with the appropriate buckets. NOTE: we are talking about ways to release the fingerprints of all the TV we archive so anybody can "search" TV for copies of sounds.