pumpp.feature.Mel

class pumpp.feature.Mel(name, sr, hop_length, n_fft, n_mels, fmax=None, log=False, conv=None)

Mel spectra feature extraction

Attributes:

name : str or None

naming scope for this feature extractor

sr : number > 0

Sampling rate of the audio (in Hz)

hop_length : int > 0

Number of samples to advance between frames

n_fft : int > 0

Number of samples per frame

n_mels : int > 0

Number of Mel frequency bins

fmax : number > 0

The maximum frequency bin. Defaults to 0.5 * sr

log : bool

If True, scale magnitude in decibels.

Otherwise, use a linear amplitude scale.

__init__(name, sr, hop_length, n_fft, n_mels, fmax=None, log=False, conv=None)

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__(name, sr, hop_length, n_fft, n_mels)	Initialize self.
layers()	Construct Keras input layers for the given transformer
merge(data)	Merge an array of output dictionaries into a single dictionary with properly scoped names.
n_frames(duration)	Get the number of frames for a given duration
phase_diff(phase)	Compute the phase differential along a given axis
pop(field)
register(key, dimension, dtype[, channels])	Register a field as a tensor with specified shape and type.
scope(key)	Apply the name scope to a key
transform(y, sr)	Transform an audio signal
transform_audio(y)	Compute the Mel spectrogram

Attributes

idx