espnet2.enh.layers.wpe.get_power

source

espnet2.enh.layers.wpe.get_power(signal, dim=-2) → Tensor

Calculates power for a given signal.

The power is calculated as the mean of the squared real and imaginary parts of the signal across the specified dimension.

• Parameters:
  • signal (Union *[*torch.Tensor , ComplexTensor ]) – Single frequency signal with shape (F, C, T), where F is the number of frequency bins, C is the number of channels, and T is the number of time steps.
  • dim (int , optional) – The dimension along which to compute the mean. Defaults to -2.
• Returns: Power with shape (F, T), representing the average power of the signal across the specified dimension.
• Return type: torch.Tensor

Examples

>>> signal = torch.tensor([[[1, 2, 3], [4, 5, 6]], [[1, 1, 1], [1, 1, 1]]])
>>> power = get_power(signal)
>>> print(power)
tensor([[  4.5000,   5.5000,   6.5000],
        [  1.0000,   1.0000,   1.0000]])