scoreur / cqt

Two implement for Constant Q Transform

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

cqt

Two implements for Constant Q Transform

SciPy(for sparse matrix representation) and NumPy are required.

API

  • fmin: minimum frequecy;
  • fmax: maximum frequecy;
  • bins: number of bins;
  • fs: the rate of sample;
  • wnd: the window function;

Customizing window function

The window function wnd receives a parameter N representing the bandwith of window and returns a numpy.array representing the corresponding weight for each position.

FAQ

Question: why does it come out ValueError: matrices are not aligned? Answer: the length of sample is small. Padding some 0's to the end of the sample can fix this.

About

Two implement for Constant Q Transform


Languages

Language:Python 100.0%