Audfprint fingerprint file
From Just Solve the File Format Problem
(Difference between revisions)
| Line 16: | Line 16: | ||
== Links == | == Links == | ||
* [https://github.com/dpwe/audfprint GitHub project] (Python) | * [https://github.com/dpwe/audfprint GitHub project] (Python) | ||
| − | * https:// | + | * https://www.ee.columbia.edu/~dpwe/LabROSA/matlab/audfprint/ - MATLAB version |
* [https://archive.org/post/1027794/new-music-analysis-files Internet Archive announcement] | * [https://archive.org/post/1027794/new-music-analysis-files Internet Archive announcement] | ||
Latest revision as of 23:06, 26 December 2024
audfprint is an audio fingerprinting utility, with native .afpt (precomputed fingerprint) and .afpk (precomputed fingerprint / "just the peaks", also called Columbia Peaks) file formats.
It was developed by Daniel Ellis at Columbia University. It was originally a MATLAB script. Newer versions are written in Python.
[edit] Identification
Files begin with ASCII "audfprinthash", or "audfprintpeak".
[edit] Sample files
- Many items at the Internet Archive Audio Archive include an .afpk file. Examples: [1], [2]
[edit] Links
- GitHub project (Python)
- https://www.ee.columbia.edu/~dpwe/LabROSA/matlab/audfprint/ - MATLAB version
- Internet Archive announcement
