mfcc#

diffsptk.MFCC#: alias of MelFrequencyCepstralCoefficientsAnalysis

class diffsptk.MelFrequencyCepstralCoefficientsAnalysis(*, fft_length: int, mfcc_order: int, n_channel: int, sample_rate: int, lifter: int = 1, f_min: float = 0, f_max: float | None = None, floor: float = 1e-05, out_format: str | int = 'y')[source]#

See this page for details.

Parameters:

fft_lengthint >= 2: The number of FFT bins, \(L\).
mfcc_orderint >= 1: The order of the MFCC, \(M\).
n_channelint >= 1: The number of mel filter banks, \(C\).
sample_rateint >= 1: The sample rate in Hz.
lifterint >= 1: The liftering coefficient.
f_minfloat >= 0: The minimum frequency in Hz.
f_maxfloat <= sample_rate // 2: The maximum frequency in Hz.
floorfloat > 0: The minimum mel filter bank output in linear scale.
out_format[‘y’, ‘yE’, ‘yc’, ‘ycE’]: y is MFCC, c is C0, and E is energy.

References

[1]

Young et al., “The HTK Book,” Cambridge University Press, 2006.

forward(x: Tensor) → Tensor[source]#

Compute the MFCC from the power spectrum.

Parameters:

xTensor [shape=(…, L/2+1)]: The power spectrum.

Returns:

yTensor [shape=(…, M)]: The MFCC without C0.
ETensor [shape=(…, 1)] (optional): The energy.
cTensor [shape=(…, 1)] (optional): The C0.

Examples

>>> x = diffsptk.ramp(19)
>>> stft = diffsptk.STFT(frame_length=10, frame_period=10, fft_length=32)
>>> mfcc = diffsptk.MFCC(
...     fft_length=32, mfcc_order=4, n_channel=8, sample_rate=8000
... )
>>> y = mfcc(stft(x))
>>> y
tensor([[-7.7745e-03, -1.4447e-02,  1.6157e-02,  1.1069e-03],
        [ 2.8049e+00, -1.6257e+00, -2.3566e-02,  1.2804e-01]])

diffsptk.functional.mfcc(x: Tensor, mfcc_order: int, n_channel: int, sample_rate: int, lifter: int = 1, f_min: float = 0, f_max: float | None = None, floor: float = 1e-05, out_format: str = 'y') → Tensor[source]#

Compute the MFCC from the power spectrum.

Parameters:

xTensor [shape=(…, L/2+1)]: The power spectrum.
mfcc_orderint >= 1: The order of the MFCC, \(M\).
n_channelint >= 1: The number of mel filter banks, \(C\).
sample_rateint >= 1: The sample rate in Hz.
lifterint >= 1: The liftering coefficient.
f_minfloat >= 0: The minimum frequency in Hz.
f_maxfloat <= sample_rate // 2: The maximum frequency in Hz.
floorfloat > 0: The minimum mel filter bank output in linear scale.
out_format[‘y’, ‘yE’, ‘yc’, ‘ycE’]: y is MFCC, c is C0, and E is energy.

Returns:

yTensor [shape=(…, M)]: The MFCC without C0.
ETensor [shape=(…, 1)] (optional): The energy.
cTensor [shape=(…, 1)] (optional): The C0.