BPM Analysis
Hi,
This is somewhat of a longshot... But I'm a computer science student and I'm currently creating a project that is similar to beaTunes, in that it is an audio application that has the ability to detect the BPM of songs. I am however having difficulty with the BPM detection part. I was wondering if you could give any info on the method you use for it?
Is it a BPM detection algorithm? Do you try and get the BPM from meta data or online sources? Or a combination?
Thanks in advance for any help you can provide.
Keyboard shortcuts
Generic
| ? | Show this help |
|---|---|
| ESC | Blurs the current field |
Comment Form
| r | Focus the comment reply box |
|---|---|
| ^ + ↩ | Submit the comment |
You can use Command ⌘ instead of Control ^ on Mac
Support Staff 1 Posted by hendrik on 06 Jul, 2022 06:45 PM
Hi,
this is not what beaTunes does, but it's a good starting point: https://scholar.google.com/citations?view_op=view_citation&hl=d...
You might also want to check https://www.tagtraum.com/tempo_estimation.html
Good luck!
-hendrik
2 Posted by William Chamber... on 07 Jul, 2022 12:58 PM
Ah amazing! I had actually cited that paper in my literature review for my thesis. I didn’t realise you were also the developer of beaTunes!
The binary for the ISMIR 2017 paper, is the source code available for that?
I had a go at installing the tempocnn python framework but had issues with conflicting dependencies. Have you encountered this before?
ERROR: Cannot install tempocnn==0.0.5 and tempocnn==0.0.6 because these package versions have conflicting dependencies.
The conflict is caused by:
tempocnn 0.0.6 depends on tensorflow==1.15.4
tempocnn 0.0.5 depends on tensorflow==1.15.4
I was using a newer version of python (3.10) than listed on the GitHub page however.
On a side note, the application I’m writing is in C++ using the JUCE framework. Do you have any advice as to how I could incorporate an existing BPM algorithm into this, whether it be python or Java, or would I need to roll my own in C++.
Thanks again for the information you’ve already provided.
Kind Regards,
William.