MFCCdeskriptorit

MFCCdeskriptorit, or Mel-Frequency Cepstral Coefficients, are a set of features widely used in the field of audio signal processing and speech recognition. They represent a compact and effective way to describe the spectral characteristics of a sound. The process of extracting MFCCs involves several steps, starting with the audio signal itself.

First, the signal is divided into short, overlapping frames. For each frame, a Fast Fourier Transform (FFT) is applied to convert the time-domain signal into its frequency-domain representation. This reveals the distribution of energy across different frequencies. Next, these frequency components are mapped onto the Mel scale, a perceptual scale of pitches judged by listeners to be equal in distance from one another. This scale is non-linear and more closely approximates human hearing by emphasizing lower frequencies.

After the Mel filter bank is applied, the logarithm of the Mel-scaled power spectrum is taken. This step is crucial because the human ear perceives loudness on a logarithmic scale. Finally, a Discrete Cosine Transform (DCT) is applied to the log-Mel power spectrum. This decorrelates the spectral components and produces a set of coefficients, the MFCCs. Typically, a small number of these coefficients (often 13 to 40) are retained, making them a computationally efficient representation of the audio. These coefficients capture the essential timbral and phonetic information of the sound, making them valuable for tasks such as speaker identification, music genre classification, and speech synthesis.