berneezy3/weightedAutocorrelation
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|
Repository files navigation
Weighted Autocorrelation function for pitch detection Pros: - robust against noise - good for speech Cons: - inaccurate for high frequencies Dependencies: scipy numpy pyaudio matplotlib scipy wave * to install these, run commands "pip scipy", "pip numpy" or "pip pyaudio" etc. Inputs: wavFile: wavFile object in python, either mono or stereo mmin: minimum pitch you expect in hertz mmax: maximum pitch you expect in hertz N: size of sliding window threshold: threshold to categorize as pitch point (play around with this to achieve better results) Output: Array object containing the pitches as floats