Fingerprint Errors, and a couple of Questions

Tags: #<Tag:0x00007f2a5695b8c0> #<Tag:0x00007f2a5695b730>


First, I want to say thank you for MusicBrainz Picard!

Second, I have a question:

I am using MusicBrainz Picard v2.1.3 and I have roughly 60,000 Songs that MusicBrainz Picard Is slowly crunching through.

In looking at the logs, however, I see many, many hundreds of lines like the errors below. What is causing these errors?


Third, when I started the process, I selected “Ignore Existing AcousticID fingerprints.” I selected “ignore” because I thought that to mean: “If your file has an acousticID, Picard will not request one.” Is that what this setting does? If not, what does this setting do?

And finally, I would like to make a donation to the cause. Where can I do that?

This indicates an error when calculating the fingerprint for a file using fpcalc. The exact reason can be different. Maybe fpcalc does not support the file format, or it cannot read the file due to some file corruption. If you want to know exactly which file is affected try starting Picard with the --debug command line argument or change the log level to debug in the log view. Then the error output will contain some more details such as:

D: 20:56:16,157 /usr/lib/python3.7/site-packages/picard/acoustid/__init__._run_next_task:213: Starting fingerprint calculator '/usr/bin/fpcalc' '/home/phw/devel/musicbrainz/picard-test-files/test.aiff'
E: 20:56:16,202 /usr/lib/python3.7/site-packages/picard/acoustid/__init__._on_fpcalc_finished:184: Fingerprint calculator failed exit code = 2, exit status = 0, error = Unknown error
D: 20:56:16,204 /usr/lib/python3.7/site-packages/picard/acoustid/__init__._lookup_fingerprint:123: AcoustID: lookup returned no result for file '/home/phw/devel/musicbrainz/picard-test-files/test.aiff'

Not quite. Getting an AcoustID for your files requires basically two steps. First it calculates an acoustic fingerprint for the audio using the fpcalc utility. Then this fingerprint gets send to the AcoustID server, which will return a matching AcoustID.

Now because calculating the fingerprint is a somewhat expensive task once the fingerprint got calculated it can be stored in the file metadata. That way tools like Picard can skip the calculation and use the fingerprint directly to lookup the AcoustID. The option “Ignore Existing AcousticID fingerprints” will ignore any fingerprints stored in the files, thus it will force recalculation of the fingerprint using fpcalc. So you assumed setting this option would save some work, while the opposite is true :wink:

Great you want to support the projects. Note that MusicBrainz and AcoustID, while closely integrated, are two separate projects. MusicBrainz is operated by the non-profit MetaBrainz foundation. AcoustID is the project of one developer, Lukáš Lalinský aka @lukz, who also did a lot of development for MusicBrainz (including a lot of the initial work on the current Picard).

For donations to AcoustID see
For donations to MetaBrainz see

And of course another great way to give back is to actually edit data on MusicBrainz and submitting correctly tagged fingerprints to AcoustID :wink: