MFCC
Definition
Mel-Frequency Cepstral Coefficients. The dominant acoustic feature set for automatic speaker recognition.
- Full name
- Mel-Frequency Cepstral Coefficients
- Primary use
- Automatic speaker recognition
- Status
- Dominant acoustic feature set in the field
Common questions
Why is the mel scale used instead of a linear frequency scale for these coefficients?+
The mel scale approximates how human hearing perceives pitch differences, giving finer resolution at lower frequencies where speech carries more distinguishing information, which makes MFCCs more effective for capturing speaker-relevant characteristics than a linear frequency representation.
Are MFCCs alone sufficient for forensic speaker comparison in court?+
MFCCs are typically one component within an automatic speaker recognition system rather than a standalone identification method, and forensic voice comparison guidelines generally call for combining automatic scores with auditory and acoustic-phonetic analysis rather than relying on MFCCs in isolation.
Related terms
- FASR
- Forensic Automatic Speaker Recognition. Likelihood-ratio framework using statistical speaker models (GMM, i-vectors, x-vectors).
- Formants (F1, F2, F3, F4)
- Resonant frequencies of the vocal tract that shape vowel quality. F1 and F2 together distinguish vowels (for example /i/ has low F1...
- Fundamental Frequency (F0)
- Rate of vocal-fold vibration during voiced speech, perceived as pitch. Typical adult male 80 to 180 Hz, adult female 165 to 255...
- Likelihood Ratio (LR)
- The ratio of two conditional probabilities: the probability of the observed evidence given the prosecution's hypothesis (same source), divided by the probability...
- Source-Filter Model
- Standard model of voice production: the glottal source (vocal-fold vibration) is filtered by the vocal-tract cavity to produce the speech signal.
- Voice Spectrogram (Sonogram)
- Time-frequency plot of speech. Wideband (about 300 Hz analysis bandwidth) shows formants; narrowband (about 45 Hz) shows harmonics.
- Voiceprint
- Lawrence Kersta's 1962 term for the claim that the spectrographic pattern of an individual's speech is unique enough to serve as a...