Post reply

Warning: this topic has not been posted in for at least 120 days.
Unless you're sure you want to reply, please consider starting a new topic.

Note: this post will not display until it's been approved by a moderator.

Name:
Email:
Subject:
Message icon:

Verification:
Sum of two plus two?:

shortcuts: hit alt+s to submit/post or alt+p to preview


Topic Summary

Posted by: Alexander K.
« on: September 12, 2019, 17:29:20 »

Not bad, guys, not bad. And in less than two years time. Thank you so much!
Posted by: Admin
« on: September 10, 2019, 21:00:30 »

We several times upgraded max.frequncy finding algorithm, current results:

Posted by: Alexander K.
« on: October 12, 2017, 19:06:25 »

Hi,
I've noticed extremely low max frequency value for some music tracks:



Spectrum analysis looks pretty good though:



Must be a bug of some sort. I've uploaded a few such tracks, which max frequencies I believe are wrong, to the Google Drive:

https://drive.google.com/open?id=0B1qgKcXZIlhRV0JDbHplMmRJb0E

Hope this'll help. Needless to say, having flows in the max frequency detection algorithm leads to lower overall file ratings, and therefore - to removal of the good tracks instead of the bad ones. I hope you'll fix the algorithm in the next release.

Windows 10 (1703) x64, Similarity 2.3.1, "classic" music analysis algorithm (not the new global optimization algorithm introduced in 2.3.1) without OpenCL acceleration.