ORCIDGitHubLinkedIn
RSS Curriculum Vitae TR

MFCC

Turkish equivalent: Mel frekansı kepstral katsayılarıDomain: Speech Processing

Mel-frequency cepstral coefficients, compact speech features derived from log Mel filter-bank energies through a cosine transform.

Speech Processing Context

A typical MFCC pipeline uses framing and windowing, an FFT-derived power spectrum, a Mel filter bank, logarithmic compression, and a Discrete Cosine Transform; some pipelines also apply pre-emphasis or energy-related terms. MFCCs became a standard compact representation in classical speech and speaker-recognition systems.

Representation Boundary

MFCC is not a lossless representation of the waveform. Phase and detailed spectral structure are discarded; the objective is to summarize short-time spectral-envelope information in a compact feature space.

Updated:

Related concepts

Speech Processing

Related technical publications

Publications whose title or summary directly references this concept.

Speech Feature Extraction with MFCC

MFCC reduces a short-time speech spectrum to a compact coefficient vector through Mel scaling, logarithmic energy, and the DCT; framing and filter-bank choices directly affect feature stability.

Audio Feature Vectors and Matching

Audio feature extraction through pre-emphasis, framing, windowing, FFT/STFT, time and spectral features, MFCCs, pitch and formants, and learned audio or speaker embeddings.