espnet2.layers.augmentation.preemphasis

source

espnet2.layers.augmentation.preemphasis(waveform, sample_rate: int, coeff: float = 0.97)

Pre-emphasize a waveform along the time dimension.

The pre-emphasis operation is defined as: : y[i] = x[i] - coeff * x[i - 1]

This process enhances the high-frequency components of the audio signal, which can be beneficial for certain types of processing, such as speech recognition.

• Parameters:
  • waveform (torch.Tensor) – audio signal (…, time)
  • sample_rate (int) – sampling rate in Hz (not used)
  • coeff (float) – pre-emphasis coefficient. Typically between 0.0 and 1.0.
• Returns: pre-emphasized signal (…, time)
• Return type: ret (torch.Tensor)

Examples

>>> import torch
>>> waveform = torch.tensor([0.0, 1.0, 0.5, 0.2])
>>> sample_rate = 16000
>>> preemphasized_waveform = preemphasis(waveform, sample_rate)
>>> print(preemphasized_waveform)