How to Plot Mfcc in Python Using Matplotlib?

Mel-frequency cepstral coefficients (MFCC) are widely used in audio signal processing and speech recognition tasks. They represent the spectral characteristics of an audio signal and are commonly used as features for various machine-learning applications. In this article, we will explore how to compute and visualize MFCC using Python and Matplotlib.

Plot Mfcc in Python Using Matplotlib

Below is the step-by-step approach to plot Mfcc in Python using Matplotlib:

Step 1: Installation

Before starting, install the following libraries with the help of the following commands:

pip install numpy librosa matplotlib

Step 2: Getting Started

Let’s start by importing the necessary libraries and loading an audio file. For this example, we’ll use the librosa library to load an audio file.

Python3

import librosa
import librosa.display
import matplotlib.pyplot as plt
import numpy as np

Step 3: Load an audio file

Now, we will load an audio file by using the following code.

Python3

audio_file = 'myaudio.wav'
y, sr = librosa.load(audio_file)

Step 4: Compute MFCC

Now, let’s compute the MFCC of the audio signal using librosa’s mfcc function:

Python3

mfccs = librosa.feature.mfcc(y, sr=sr, n_mfcc=13)

Step 5: Visualize MFCC

To visualize the MFCC, we can use Matplotlib to create a heatmap. Each row in the MFCC matrix represents a different coefficient, and each column represents a frame in the audio signal.

Python3

plt.figure(figsize=(10, 5))
librosa.display.specshow(mfccs, x_axis='time')
plt.colorbar()
plt.title('MFCC')
plt.xlabel('Time')
plt.ylabel('MFCC Coefficients')
plt.show()

Step 6: Customization

You can customize the plot further by adding axis labels, title, and adjusting the color map. Additionally, you can choose the number of MFCC coefficients by modifying the n_mfcc parameter when computing the MFCC.

Python3

plt.figure(figsize=(10, 5))
librosa.display.specshow(mfccs, x_axis='time', cmap='viridis', hop_length=512)
plt.colorbar(format='%+2.0f dB')
plt.title('MFCC')
plt.xlabel('Time (s)')
plt.ylabel('MFCC Coefficients')
plt.show()

Complete Code

Python3

import librosa
import librosa.display
import matplotlib.pyplot as plt

def plot_mfcc(audio_path):
    # Load the audio file
    y, sr = librosa.load(audio_path)

    # Extract MFCC features
    mfccs = librosa.feature.mfcc(y=y, sr=sr, n_mfcc=13)

    # Plot MFCCs
    plt.figure(figsize=(10, 4))
    librosa.display.specshow(mfccs, x_axis='time', cmap='viridis')
    plt.colorbar(format='%+2.0f dB')
    plt.title('MFCC')
    plt.xlabel('Time')
    plt.ylabel('MFCC Coefficient')
    plt.show()

# Example usage
audio_file_path = "path/to/your/audio/file.wav"
plot_mfcc(audio_file_path)

Output: