You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is Crepe basically trained on vocals (speeches)? I tried myself to track pitches on separate vocals and instrumental tracks of the same song and it resulted in considerably high pitch differences at the same time points for vocals and instrumental.
The text was updated successfully, but these errors were encountered:
I'm not sure I fully understand the question, but, CREPE was trained on a combination of speech and other instrument sounds, and is designed to work on any pitched sound (speech or otherwise) as long as it's monophonic (i.e. single voice/instrument, no accompaniment). It will not work (as expected) on mixtures of instruments.
Is Crepe basically trained on vocals (speeches)? I tried myself to track pitches on separate vocals and instrumental tracks of the same song and it resulted in considerably high pitch differences at the same time points for vocals and instrumental.
The text was updated successfully, but these errors were encountered: