| | 93 | |
| | 94 | |
| | 95 | == Audio Fingerprinting Filters for FFmpeg == |
| | 96 | |
| | 97 | '''Description:''' Implementing filters for ffmpeg that would give different audio |
| | 98 | fingerprints for an audio which could be reused by other people for variety |
| | 99 | of applications. Goal of this system is given a song as an input, it would |
| | 100 | spits out similar sounding songs. Some of its applications are: |
| | 101 | |
| | 102 | * detecting duplicate songs |
| | 103 | * retrieving similar songs |
| | 104 | * can be used for recommending songs |
| | 105 | * clustering songs based on content/tags |
| | 106 | * identifying a fragment of a recording |
| | 107 | |
| | 108 | '''Expected results:''' A filter that can reliably and quickly identify a degraded song in a set of a million, better a billion songs |
| | 109 | |
| | 110 | '''Prerequisites:''' Good C coding skills, basic familiarity with git, knowledge about audio fingerprinting |
| | 111 | |
| | 112 | '''Qualification Task:''' |
| | 113 | |
| | 114 | '''Mentor:''' Michael Niedermayer (''michaelni'' in #ffmpeg-devel on Freenode IRC, michaelni@gmx.at) |
