I just ran analyze all songs and although it got better at picking some of the duplicates there are songs that it is picking out the have different names different times and BPM. When I play the songs the labels and songs are correct but they are different songs that it tells me are duplicates. Is there a way to make the filtering more reliable? A lot of the suggestions are just wrong and I have to go through every file to make sure it is correct. What else can I do to make it work better?
Thanks that helps some when looking for duplicates. Is there a way to know which is the correct track when the tracks are showing different songs but acoustically identical or is listening to the tracks the only way to be sure?
Can this program correct the track names artist and general metadata from the analyze function or is there another function to fix the wrong data reliably? I have a lot of songs that are showing the wrong artist song or many other things and need to have this fixed this is the main reason I brought this program.
hendrik on 10 Aug, 2017 11:56 AM
Is there a way to know which is the correct track when the tracks are showing different songs but acoustically identical or is listening to the tracks the only way to be sure?
You could try running the analysis with the import metadata task enabled and have it overwrite your metadata. If your songs are badly mislabeled, to get "correct" results, you would need to check the force lookup ... box.
However... The data would be retrieved from the beaTunes database—a central database filled with user submissions. That means that it's very large, but some values may be incorrect. Another issue here is, that often tracks are released on multiple albums. They are acoustically identical, but there is no way of knowing which album you want.
Another issue here is that running an analysis is a mass operation with no undo. That makes it inherently risky on a large collection (a backup is a good idea!). So in essence, your mileage may vary.