Menu
I'm using the librosa library to convert music segments into mel-spectrograms to use as inputs for my neural network, as shown in the docs here.
How is this different from MFCCs, if at all? Are there any advantages or disadvantages to using either?
782
The linear audio spectrogram is ideally suited for applications where all frequencies have equal importance, while mel spectrograms are better suited for applications that need to model human hearing perception. Mel spectrogram data is also suited for use in audio classification applications.
Since, Mel-frequency bands are distributed evenly in MFCC and they are much similar to the voice system of a human, thus, MFCC can efficiently be used to characterize speakers, for instance, it can be used to recognize the speaker's cell phone model details and further the details of the speaker.
The MFCC feature extraction technique basically includes windowing the signal, applying the DFT, taking the log of the magnitude, and then warping the frequencies on a Mel scale, followed by applying the inverse DCT. The detailed description of various steps involved in the MFCC feature extraction is explained below.
To get MFCC, compute the DCT on the mel-spectrogram. The mel-spectrogram is often log-scaled before.
MFCC is a very compressible representation, often using just 20 or 13 coefficients instead of 32-64 bands in Mel spectrogram. The MFCC is a bit more decorrelarated, which can be beneficial with linear models like Gaussian Mixture Models. With lots of data and strong classifiers like Convolutional Neural Networks, mel-spectrogram can often perform better.
157
I suppose, jonnor's answer is not exactly correct. There are two steps:
1. Take logs of Mel spectrogram.
2. Compute DCT on logs.
Moreover, taking logs seems to be "the main part" for training NN: https://qr.ae/TWtPLD
30
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With
