spec#

class diffsptk.Spectrum(fft_length: int, *, eps: float = 0, relative_floor: float | None = None, out_format: str | int = 'power')[source]#

See this page for details.

Parameters:

fft_lengthint >= 2: The number of FFT bins, \(L\).
epsfloat >= 0: A small value added to the power spectrum.
relative_floorfloat < 0 or None: The relative floor of the power spectrum in dB.
out_format[‘db’, ‘log-magnitude’, ‘magnitude’, ‘power’]: The output format.

forward(b: Tensor | None = None, a: Tensor | None = None) → Tensor[source]#

Compute spectrum.

Parameters:

bTensor [shape=(…, M+1)] or None: The numerator coefficients.
aTensor [shape=(…, N+1)] or None: The denominator coefficients.

Returns:

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

Examples

>>> x = diffsptk.ramp(1, 3)
>>> x
tensor([1., 2., 3.])
>>> spec = diffsptk.Spectrum(8)
>>> y = spec(x)
>>> y
tensor([36.0000, 25.3137,  8.0000,  2.6863,  4.0000])

diffsptk.functional.spec(b: Tensor | None = None, a: Tensor | None = None, *, fft_length: int = 512, eps: float = 0, relative_floor: float | None = None, out_format: str = 'power') → Tensor[source]#

Compute spectrum.

Parameters:

bTensor [shape=(…, M+1)] or None: The numerator coefficients.
aTensor [shape=(…, N+1)] or None: The denominator coefficients.
fft_lengthint >= 2: The number of FFT bins, \(L\).
epsfloat >= 0: A small value added to the power spectrum.
relative_floorfloat < 0 or None: The relative floor of the power spectrum in dB.
out_format[‘db’, ‘log-magnitude’, ‘magnitude’, ‘power’]: The output format.

Returns:

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