specan method from warbleR library (Librosa version?)

[Jose Arrieta]

Hi guys, 

I was checking libraries for acoustic analysis in differents languages an found specan method from warbleR, and it  measures 22 acoustic parameters on acoustic signals.

Would be cool to have a similar one for Librosa. WDY? 

• duration: length of signal 

• meanfreq: mean frequency (in kHz) 

• sd: standard deviation of frequency 

• median: median frequency (in kHz)

 • Q25: first quantile (in kHz) • Q75: third quantile (in kHz) 

• IQR: interquantile range (in kHz) 

• skew: skewness (see note in specprop description) 

• kurt: kurtosis (see note in specprop description) 

• sp.ent: spectral entropy • sfm: spectral flatness 

• mode: mode frequency • centroid: frequency centroid (see specprop) 

• peakf: peak frequency (frequency with highest energy) 

• meanfun: average of fundamental frequency measured across acoustic signal 

• minfun: minimum fundamental frequency measured across acoustic signal 

• maxfun: maximum fundamental frequency measured across acoustic signal 

• meandom: average of dominant frequency measured across acoustic signal 

• mindom: minimum of dominant frequency measured across acoustic signal 

• maxdom: maximum of dominant frequency measured across acoustic signal 

• dfrange: range of dominant frequency measured across acoustic signal 

• modindx: modulation index. Calculated as the accumulated absolute difference between adjacent measurements of dominant frequencies divided by the dominant frequency range

 • startdom: dominant frequency measurement at the start of the signal 

• enddom: dominant frequency measurement at the end of the signal

 • dfslope: slope of the change in dominant through time ([enddom-startdom]/duration)

https://cran.r-project.org/web/packages/warbleR/warbleR.pdf

[Brian McFee]

Pull requests are always welcome. :)

I would not want all of this done from a single `specan` function though; each feature should be its own function.