Select Git revision
functions.py 2.95 KiB
##### IMPORTATIONS #####
import os
import json
import numpy as np
from librosa import load, amplitude_to_db, stft, pcen
from scipy.signal import resample
from line_clicker.line_clicker import to_curve
##### FUNCTIONS #####
def save_dict(dictionary, folder, name):
"""
A function that saves a dictionary to a given path.
...
Parameters
----------
dictionary : dict
Any dictionary.
folder : str
Path to the folder where the dictionary will be saved.
name : str
Name of the file in which the dictionary will be saved.
If there is an extension, should not be different than .json extension.
Returns
-------
None : save dict to json file.
"""
if len(dictionary) > 0:
with open(os.path.join(folder, name), "w") as f:
json.dump(dictionary, f, indent=4)
def load_waveform(wavefile_name, sr_resample):
"""
A function that loads any given wavefile
and it resamples it to a given sampling rate.
...
Parameters
----------
wavefile_name : str
Path of the wavefile that will be loaded.
sr_resample : int
Resampling rate for the waveform.
Returns
-------
wavefile_dec : numpy array
Loaded and resampled waveform
"""
wavefile, sr = load(wavefile_name, sr=None)
wavefile_dec = resample(wavefile,
int(((len(wavefile)/sr)*sr_resample)))
return wavefile_dec
def wave_to_spectrogram(waveform, SR, n_fft, w_size, clip, as_pcen=False):
"""
A function that transforms any given waveform to a spectrogram.
...
Parameters
----------
waveform : numpy array
Waveform of an audio recording. Shape should be (N, 1).
SR : int
Sampling rate of the waveform
n_fft : int
Desired size for fft window. Should be in [1, N-1].
w_size : int
Desired size for hop length between two fft. Should be in [1, N-1].
clip : int
Clipping value for dB. If pixel value < clip, pixel is turned into NaN.
If as_pcen is selected, clipping will be applied using the values of pixels
in the orgiginal spectrogram.
as_pcen : bool, optional.
Whether the returned image should be a PCEN or not.
Aka : spectrogram with enhanced contrast.
Default is False.
Returns
-------
spectro : numpy array
Spectrogram of the waveform using the provided parameters.
audio_length : float
Duration of the audio in seconds.
"""
base = np.abs(stft(
waveform,
n_fft=n_fft,
hop_length=w_size))
spectro_pcen = pcen(base * (2**31), bias=10)
spectro_og = amplitude_to_db(base)
spectro_og = spectro_og - (np.max(spectro_og))
spectro_og[spectro_og < clip] = np.nan
if as_pcen:
spectro_pcen[np.isnan(spectro_og)] = np.nan
spectro = spectro_pcen
else:
spectro = spectro_og
audio_length = len(waveform)/SR
return spectro, audio_length