The Difference librosa.filters.mel() and librosa.feature.melspectrogram() – Librosa Tutorial

By | June 10, 2022
0 Comment

To compute mel in librosa, we can use librosa.filters.mel() and librosa.feature.melspectrogram(). In this tutorial, we will introduce the difference between them.

librosa.filters.mel()

It is defined as:

  1. librosa.filters.mel(*, sr, n_fft, n_mels=128, fmin=0.0, fmax=None, htk=False, norm='slaney', dtype=<class 'numpy.float32'>)
librosa.filters.mel(*, sr, n_fft, n_mels=128, fmin=0.0, fmax=None, htk=False, norm='slaney', dtype=<class 'numpy.float32'>)

It will create a Mel filter-bank and produce a linear transformation matrix to project FFT bins onto Mel-frequency bins.

Notice: It creates a Mel filter-bank does not FBank, you can not use it as audio feature.

For example:

  1. import librosa
  2. import numpy as np
  3. import matplotlib.pyplot as plt
  5. def plot_mel_fbank(fbank, title=None):
  6. fig, axs = plt.subplots(1, 1)
  7. axs.set_title(title or "Filter bank")
  8. axs.imshow(fbank, aspect="auto")
  9. axs.set_ylabel("frequency bin")
  10. axs.set_xlabel("mel bin")
  11. plt.show(block=False)
  12. sr = 8000
  14. mels = librosa.filters.mel(sr=sr, n_fft = 512, fmin=0.0, fmax=sr / 2.0,n_mels=80)
  15. print(mels.shape)
  16. plot_mel_fbank(mels)
import librosa
import numpy as np
import matplotlib.pyplot as plt

def plot_mel_fbank(fbank, title=None):
    fig, axs = plt.subplots(1, 1)
    axs.set_title(title or "Filter bank")
    axs.imshow(fbank, aspect="auto")
    axs.set_ylabel("frequency bin")
    axs.set_xlabel("mel bin")
    plt.show(block=False)
sr = 8000

mels = librosa.filters.mel(sr=sr, n_fft = 512, fmin=0.0, fmax=sr / 2.0,n_mels=80)
print(mels.shape)
plot_mel_fbank(mels)

Run this code, you will see:

  1. (80, 257)
(80, 257)

librosa.filters.mel() example

We can find this function only returns a weight, it can not process any audio data.

librosa.feature.melspectrogram()

It is defined as:

  1. librosa.feature.melspectrogram(*, y=None, sr=22050, S=None, n_fft=2048, hop_length=512, win_length=None, window='hann', center=True, pad_mode='constant', power=2.0, **kwargs)
librosa.feature.melspectrogram(*, y=None, sr=22050, S=None, n_fft=2048, hop_length=512, win_length=None, window='hann', center=True, pad_mode='constant', power=2.0, **kwargs)

It can compute a mel-scaled spectrogram.

Notice: The result of this function can be used as the feature of an audio.

In order to understand how to use this function, you can read:

Compute Audio Log Mel Spectrogram Feature: A Step Guide – Python Audio Processing

We will use an example to show the effect of mel spectrogram.

  1. import librosa
  2. import numpy as np
  3. import matplotlib.pyplot as plt
  5. def plot_spectrogram(spec, title=None, ylabel="freq_bin", aspect="auto", xmax=None):
  6. fig, axs = plt.subplots(1, 1)
  7. axs.set_title(title or "Spectrogram (db)")
  8. axs.set_ylabel(ylabel)
  9. axs.set_xlabel("frame")
  10. im = axs.imshow(librosa.power_to_db(spec), origin="lower", aspect=aspect)
  11. if xmax:
  12. axs.set_xlim((0, xmax))
  13. fig.colorbar(im, ax=axs)
  14. plt.show(block=False)
  16. audio_file = 'speech-01-002.flac'
  17. sr = 8000
  18. audio_data, sr = librosa.load(audio_file, sr= sr, mono=True)
  19. print(audio_data.shape)
  20. win_length = int(0.025 * sr)
  21. hop_length = int(0.01 * sr)
  22. melspectrum = librosa.feature.melspectrogram(y=audio_data, sr=sr, hop_length= hop_length, win_length = win_length, window='hann', n_fft = 512, n_mels=80)
  23. print(melspectrum.shape)
  24. plot_spectrogram(melspectrum)
import librosa
import numpy as np
import matplotlib.pyplot as plt

def plot_spectrogram(spec, title=None, ylabel="freq_bin", aspect="auto", xmax=None):
    fig, axs = plt.subplots(1, 1)
    axs.set_title(title or "Spectrogram (db)")
    axs.set_ylabel(ylabel)
    axs.set_xlabel("frame")
    im = axs.imshow(librosa.power_to_db(spec), origin="lower", aspect=aspect)
    if xmax:
        axs.set_xlim((0, xmax))
    fig.colorbar(im, ax=axs)
    plt.show(block=False)

audio_file = 'speech-01-002.flac'
sr = 8000
audio_data, sr = librosa.load(audio_file, sr= sr, mono=True)
print(audio_data.shape)
win_length = int(0.025 * sr)
hop_length = int(0.01 * sr)
melspectrum = librosa.feature.melspectrogram(y=audio_data, sr=sr, hop_length= hop_length, win_length = win_length, window='hann', n_fft = 512, n_mels=80)
print(melspectrum.shape)
plot_spectrogram(melspectrum)

Run this code, you will see:

  1. (389724,)
  2. (80, 4872)
(389724,)
(80, 4872)

librosa.feature.melspectrogram() example

For the source code of librosa.feature.melspectrogram(), we can find:

  1. # Build a Mel filter
  2. mel_basis = filters.mel(sr=sr, n_fft=n_fft, **kwargs)
  4. return np.einsum("...ft,mf->...mt", S, mel_basis, optimize=True)
    # Build a Mel filter
    mel_basis = filters.mel(sr=sr, n_fft=n_fft, **kwargs)

    return np.einsum("...ft,mf->...mt", S, mel_basis, optimize=True)

librosa.filters.mel() is used in librosa.feature.melspectrogram().

Leave a Reply