Le nonidi 19 vendémiaire, an CCXXV, Pallavi Kumari a écrit : > Implementing filters for ffmpeg that would give different audio > fingerprints for an audio which could be reused by other people for variety > of applications. Goal of this system is given a song as an input, it would > spits out similar sounding songs. Some of its applications are: > > * detecting duplicate songs > * retrieving similar songs > * can be used for recommending songs > * clustering songs based on content/tags
Sorry I did not follow up earlier in the discussion: maybe you could consider adding: * identifying a fragment of a recording I do not think this is covered by the other cases. My idea is this: imagine you have several recordings of some event (maybe a speech, or a concert), and you want to identify a small portion for something, a comment for example. If there were a reference recording available for anyone, you could just give the timing. But if not, you could identify the portion by its fingerprint. Then anyone who has a recording can look for the portion using the finger print. Another use case would be subtitles. It could look like catering for piracy, since official sources of subtitles certainly have all the timing information they need. But I can see legitimate reasons to take subtitles from the web even with a legal copy of the program: maybe living in a foreign country where the TV broadcasts do not have subtitles in the appropriate language. Regards, -- Nicolas George
signature.asc
Description: Digital signature
_______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
