espnet2.layers.augmentation.reverse

source

espnet2.layers.augmentation.reverse(waveform, sample_rate)

Reverse the input waveform along the time dimension.

• Parameters:
  • waveform (torch.Tensor) – audio signal (…, time).
  • sample_rate (int) – sampling rate in Hz (not used).
• Returns: the reversed audio signal (…, time).
• Return type: torch.Tensor

Examples

>>> waveform = torch.tensor([1.0, 2.0, 3.0])
>>> reversed_waveform = reverse(waveform, sample_rate=16000)
>>> print(reversed_waveform)
tensor([3.0, 2.0, 1.0])