import librosa
import numpy as np
[docs]def from_mel(
mel_,
sr=16000,
n_fft=2048,
n_iter=32,
win_length=1000,
hop_length=100,
):
"""
Change melspectrogram into waveform using Librosa.
Parameters
----------
spectrogram: np.array
Returns
--------
result: np.array
"""
return librosa.feature.inverse.mel_to_audio(
mel_,
sr=sr,
n_fft=n_fft,
hop_length=hop_length,
win_length=win_length,
window='hann',
center=True,
pad_mode='reflect',
power=1.0,
n_iter=n_iter,
)
[docs]def from_mel_vocoder(
mel,
sr=22050,
n_fft=1024,
n_mels=256,
fmin=80,
fmax=7600,
n_iter=32,
win_length=None,
hop_length=256,
):
"""
Change melspectrogram into waveform using Librosa.
Parameters
----------
spectrogram: np.array
Returns
--------
result: np.array
"""
mel_spec = np.power(10.0, mel)
mel_basis = librosa.filters.mel(
sr, n_fft=n_fft, n_mels=n_mels, fmin=fmin, fmax=fmax
)
mel_to_linear = np.maximum(
1e-10, np.dot(np.linalg.pinv(mel_basis), mel_spec)
)
gl_lb = librosa.griffinlim(
mel_to_linear,
n_iter=n_iter,
hop_length=hop_length,
win_length=win_length,
)
return gl_lb
