If features are highly correlate with each other, we can solve this correlation by rotating the axis, by PCA.
Gaussian distribution of classification result of feature vector
We now start thinking about what a good representation of the acoustic signal should be, motivating the use Mel-Frequency Cepstral Coefficients (MFCCs).
Since the feature in a feature vector is correlated, if we want to use gaussian, we have to solve this correlation problem.
In ASR, we can decompose the input further
The independent variable of a cepstral graph is called the quefrency.
The quefrency is a measure of time, though not in the sense of a signal in the time domain.
Then the first 12 value in cepstrum is our result feature vector, after feature engineering.
Then we can solve the correlation problem discussed in previous, by using MFCCs.
Overview of steps required to derive MFCCs, moving towards modelling MFCCs with Gaussians and Hidden Markov Models
Origin: Module 8 – Speech Recognition – Feature engineering
Translate + Edit: YangSier (Homepage)