Skip to content

Soundable-EDA is an exploratory data analysis pipeline for deconvolution of audio samples as a precursor to feature extracting with supervised machine larning.

Notifications You must be signed in to change notification settings

DataSperling/soundable-EDA

Repository files navigation

Repo for EDA of audio samples, choosing FFT / DFT algorithms, dimensionality reduction and feature selection.

FFT_wave_file.py: A python script for determining the applicability of basic scipy.fft and scipy.fftfreq methods for the analysis and feature extraction in real-world (high-white-noise) sound scenes

FFT-xeno-canto-82416.ipynb: jupyter workbook showing the filtering and FFT of a real-world audio sample

About

Soundable-EDA is an exploratory data analysis pipeline for deconvolution of audio samples as a precursor to feature extracting with supervised machine larning.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages