What is the difference between all the BPM plugins? Which one is the best?

[shimizurei]

I just want an accurate BPM for the BPM field.

Looking at the docs there is:

• https://github.com/adamjakab/BeetsPluginBpmAnalyser
• https://beets.readthedocs.io/en/stable/plugins/bpm.html
• https://beets.readthedocs.io/en/latest/plugins/autobpm.html

I'm working on my config file (haven't used Beets in years) and I have no idea which one to use. I just know I won't be using the keystroke one because I'd mess that up.

[adamjakab]

Hi @shimizurei,

The second one bpm: "lets you to get the tempo (beats per minute) of a song by tapping out the beat on your keyboard." . Even though I think it is a good exercise probably a time consuming choice if you have Gigs of music to go through.

I am the author of the first one BeetsPluginBpmAnalyser (so I might be biased ;) ) and I believe that at that time the autobpm plugin did not exist. I think that the main difference is in the music analysis library they use: aubio and librosa. I do not know if one library is more performant than the other and/or if the calculated values are more precise with aubio or with librosa.

I also see some slight differences in the configuration options where BeetsPluginBpmAnalyser allow for a more fine-tuned configuration:

bpmanalyser:
  auto: no
  dry-run: no
  write: yes
  threads: 2
  force: no
  quiet: no

Maybe you could install both and run a comparison. That could be interesting.

EDIT: Pls note that (probably due to some changes in the aubio library) atm the BeetsPluginBpmAnalyser plugin fails with error; RuntimeError: AUBIO ERROR: source_wavread: Failed opening /mnt/Music/xyz.mp3 (could not find RIFF header).
I just found this thread: aubio/aubio#116 which reports the same issue.
I will look into it and get it fixed.

EDIT2: I fixed the plugin. I now uses the pydub library (which relies on ffmpeg or libav) to convert the audio files to wav before analyzing them with aubio.

I hope this helps.