ap#

class diffsptk.Aperiodicity(frame_period, sample_rate, fft_length, algorithm='tandem', out_format='a', **kwargs)[source]#

See this page for details.

Parameters:

frame_periodint >= 1: Frame period, \(P\).
sample_rateint >= 1: Sample rate in Hz.
fft_lengthint: Size of double-sided aperiodicity, \(L\).
algorithm[‘tandem’]: Algorithm.
out_format[‘a’, ‘p’, ‘a/p’, ‘p/a’]: Output format.
lower_boundfloat >= 0: Lower bound of aperiodicity.
upper_boundfloat <= 1: Upper bound of aperiodicity.
window_length_msint >= 1: Window length in msec.
epsfloat > 0: A number used to stabilize colesky decomposition.

References

[1]

H. Kawahara et al., “Simplification and extension of non-periodic excitation source representations for high-quality speech manipulation systems,” Proceedings of Interspeech, pp. 38-41, 2010.

forward(x, f0)[source]#

Compute aperiodicity measure.

Parameters:

xTensor [shape=(B, T) or (T,)]: Waveform.
f0Tensor [shape=(B, N) or (N,)]: F0 in Hz.

Returns:

outTensor [shape=(B, N, L/2+1) or (N, L/2+1)]: Aperiodicity.

Examples

>>> x = diffsptk.sin(100, 10)
>>> pitch = diffsptk.Pitch(80, 16000, out_format="f0")
>>> f0 = pitch(x)
>>> f0
tensor([1597.2064, 1597.2064])
>>> aperiodicity = diffsptk.Aperiodicity(80, 16000, 8)
>>> ap = aperiodicity(x, f0)
>>> ap
tensor([[0.0010, 0.0010, 0.1729, 0.1647, 0.1569],
        [0.0010, 0.0010, 0.0490, 0.0487, 0.0483]])

ap#

This Page