carat.features.melSpectrogram

carat.features.melSpectrogram(in_spec, in_time, in_freq, nfilts=40, minfreq=20, maxfreq=None)

This function converts a Spectrogram with linearly spaced frequency components to the Mel scale.

Given an input signal, it calculates the DFT of frames of the signal and stores them in bi-dimensional Scipy array.

Args:

• window_len (float): length of the window in seconds (must be positive).
• window (callable): a callable object that receives the window length in samples and returns a numpy array containing the windowing function samples.

Returns:

• spec (numpy array): mel-spectrogram data
• time (numpy array): time in seconds of each frame
• frequnecy (numpy array): frequency grid