Skip to content

AI-Associates/knime-dsp-nodes

Repository files navigation

Project Infos

This project contains the KNIME nodes or digital signal processing.

Fast Fourier Transform (FFT)

This node performs a forward Fast Fourier Transformation(FFT) on each row of the input table. Generally speaking, it extracts the frequencies of an input signal.

Frequency Domain Features (FDF)

This node computes statistics in the frequency domain. The chosen statistics will be computed for each row.

Time Domain Features (TDF)

This node computes statistics in the time domain. The chosen statistics will be computed for each row.

WAV Reader

This node reads in .wav files.

Welch Averaging

This node bundles together a number of rows and takes an average in every column for this bundle.

Window Function

This node multiplies the input rows with a window function to prevent spectral leakage in subsequent FFT.

Window Slider

This node creates sliding windows from an input signal which is stored in a table with one column.