A distributor receives an audio file.<\/p>\n\n\n\n
How could it possibly know that part of the recording came from somebody else’s song?<\/p>\n\n\n\n
One answer is audio fingerprinting.<\/p>\n\n\n\n
A fingerprinting system creates a mathematical representation of characteristics within a recording.<\/p>\n\n\n\n
That fingerprint can then be compared with reference recordings.<\/p>\n\n\n\n
It is conceptually similar to recognising a fingerprint rather than comparing filenames.<\/p>\n\n\n\n
Calling a file:<\/p>\n\n\n\n
does not change the audio inside it.<\/p>\n\n\n\n If somebody uploads the same recording again, recognition technology may identify the match.<\/p>\n\n\n\n But systems are becoming more sophisticated.<\/p>\n\n\n\n Some technology can identify source material after changes including:<\/p>\n\n\n\n SoundOn, for example, announced technology from ACRCloud earlier this year specifically designed to identify derivative recordings despite substantial speed or pitch changes. <\/p>\n\n\n\n No.<\/p>\n\n\n\n This is important.<\/p>\n\n\n\n There may be legitimate reasons recordings match.<\/p>\n\n\n\n For example:<\/p>\n\n\n\n A detection system identifies similarity.<\/p>\n\n\n\n Rights still require context.<\/p>\n\n\n\n If every audio match automatically produced a rejection, legitimate releases would sometimes be blocked.<\/p>\n\n\n\n A distributor may therefore ask for:<\/p>\n\n\n\n Audio recognition makes it increasingly difficult to hide copied recordings through simple technical changes.<\/p>\n\n\n\n The correct solution to copyrighted audio is not trying to evade detection.<\/p>\n\n\n\n It is obtaining the rights.<\/p>\n","protected":false},"excerpt":{"rendered":" A distributor receives an audio file. How could it possibly know that part of the recording came from somebody else’s song? One answer is audio fingerprinting. What is audio fingerprinting? A fingerprinting system creates a mathematical representation of characteristics within a recording. That fingerprint can then be compared with reference recordings. It is conceptually similar to recognising a fingerprint rather than comparing filenames. Calling a file: my-original-song-final.wav does not change the audio inside it. Exact copies are relatively straightforward If somebody uploads the same recording again, recognition technology may identify the match. But systems are becoming more sophisticated. Altered recordings…<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[1],"tags":[1503,31],"coauthors":[2],"class_list":["post-4643","post","type-post","status-publish","format-standard","hentry","category-news","tag-audio","tag-music-distribution"],"acf":[],"yoast_head":"\nmy-original-song-final.wav<\/code><\/p>\n\n\n\nExact copies are relatively straightforward<\/h2>\n\n\n\n
Altered recordings<\/h2>\n\n\n\n
\n
Does a match prove infringement?<\/h2>\n\n\n\n
\n
Why human review matters<\/h2>\n\n\n\n
\n
Final thoughts<\/h2>\n\n\n\n