Dan Ellis

I made a submission to an audio fingerprinting evaluation in 2014, there are details in README-dpwe-audfprint.txt This includes a breakdown of the compute times between fingerprint calculation and matching. For...

I think you want “add” instead of “new” for all except the first call, and use the same dbase name for all folders. “Pklz” is the suffix for precomputed per-track...

Unfortunately not. The idea of the fingerprints is that they are very small but still able to correctly distinguish items. But you’re not going to be able to reconstruct much...

Match duration is just the outer limits of the earliest and latest matching and aligned landmarks. Match hash density (total matching hashes divided by match time) is the most consistent...

Gosh, I have zero memory of adding that function to the Matlab audfprint. No, I never added it to the python version. In principle, the same information is available, but...

The relevant Matlab code is at: https://labrosa.ee.columbia.edu/%7Edpwe/resources/matlab/audfprint/alignhashes.m.html It actually performs a brute-force search for time warps in the range -2..2% in steps of 0.1% because I couldn't find a good...

@ezavesky : Let me provide a bit more detail to see if it helps you. As you mention, it's not that rare to encounter audio that is not at exactly...

It's having a problem running ffmpeg on the mp3 file, or in parsing the results. Can you run "ffmpeg -i 1.mp3" and paste the output here? I'm running with ffmpeg...

> First, is there any way to force a higher sensitivity without requiring an > overall higher density? In other words, in the areas with lots of peaks (at >...

They're different mixes of the Ad. I see this quite a lot. In particular, the version in p1-1580186349-0609.mp3 is substantially time compressed - the voice comes in 2 seconds sooner...