Statements (27)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
feature extraction technique
|
| gptkbp:basedOn |
mel scale
|
| gptkbp:capturedBy |
short-term power spectrum of sound
|
| gptkbp:dimensions |
typically 12 or 13 coefficients
|
| gptkbp:firstPublished |
gptkb:IEEE_Transactions_on_Acoustics,_Speech,_and_Signal_Processing
1980 |
| https://www.w3.org/2000/01/rdf-schema#label |
MFCC
|
| gptkbp:input |
audio signal
|
| gptkbp:output |
feature vector
|
| gptkbp:proposedBy |
Paul Mermelstein
Steven B. Davis |
| gptkbp:relatedTo |
gptkb:LPC
gptkb:PLP spectrogram |
| gptkbp:standsFor |
Mel-Frequency Cepstral Coefficients
|
| gptkbp:step |
gptkb:Fourier_transform
gptkb:discrete_cosine_transform windowing framing logarithm mel filterbank pre-emphasis |
| gptkbp:usedIn |
speech recognition
audio signal processing speaker identification |
| gptkbp:bfsParent |
gptkb:torchaudio
|
| gptkbp:bfsLayer |
6
|