Add rolloff param to resample (#1488)

torchaudio/compliance/kaldi.py

@@ -755,7 +755,8 @@ def mfcc(
 def resample_waveform(waveform: Tensor,
                       orig_freq: float,
                       new_freq: float,
-                      lowpass_filter_width: int = 6) -> Tensor:
+                      lowpass_filter_width: int = 6,
+                      rolloff: float = 0.99) -> Tensor:
     r"""Resamples the waveform at the new frequency.
 
     This is a wrapper around ``torchaudio.functional.resample``.
@@ -766,8 +767,10 @@ def resample_waveform(waveform: Tensor,
         new_freq (float): The desired frequency
         lowpass_filter_width (int, optional): Controls the sharpness of the filter, more == sharper
             but less efficient. We suggest around 4 to 10 for normal use. (Default: ``6``)
+        rolloff (float, optional): The roll-off frequency of the filter, as a fraction of the Nyquist.
+            Lower values reduce anti-aliasing, but also reduce some of the highest frequencies. (Default: ``0.99``)
 
     Returns:
         Tensor: The waveform at the new frequency
     """
-    return torchaudio.functional.resample(waveform, orig_freq, new_freq, lowpass_filter_width)
+    return torchaudio.functional.resample(waveform, orig_freq, new_freq, lowpass_filter_width, rolloff)

torchaudio/functional/functional.py

@@ -1298,8 +1298,13 @@ def compute_kaldi_pitch(
     return result
 
 
-def _get_sinc_resample_kernel(orig_freq: int, new_freq: int, lowpass_filter_width: int,
-                              device: torch.device, dtype: torch.dtype):
+def _get_sinc_resample_kernel(
+        orig_freq: int,
+        new_freq: int,
+        lowpass_filter_width: int,
+        rolloff: float,
+        device: torch.device,
+        dtype: torch.dtype):
     assert lowpass_filter_width > 0
     kernels = []
     base_freq = min(orig_freq, new_freq)
@@ -1307,7 +1312,7 @@ def _get_sinc_resample_kernel(orig_freq: int, new_freq: int, lowpass_filter_widt
     # At first I thought I only needed this when downsampling, but when upsampling
     # you will get edge artifacts without this, as the edge is equivalent to zero padding,
     # which will add high freq artifacts.
-    base_freq *= 0.99
+    base_freq *= rolloff
 
     # The key idea of the algorithm is that x(t) can be exactly reconstructed from x[i] (tensor)
     # using the sinc interpolation formula:
@@ -1352,7 +1357,8 @@ def resample(
         waveform: Tensor,
         orig_freq: float,
         new_freq: float,
-        lowpass_filter_width: int = 6
+        lowpass_filter_width: int = 6,
+        rolloff: float = 0.99,
 ) -> Tensor:
     r"""Resamples the waveform at the new frequency. This matches Kaldi's OfflineFeatureTpl ResampleWaveform
     which uses a LinearResample (resample a signal at linearly spaced intervals to upsample/downsample
@@ -1369,6 +1375,8 @@ def resample(
         new_freq (float): The desired frequency
         lowpass_filter_width (int, optional): Controls the sharpness of the filter, more == sharper
             but less efficient. We suggest around 4 to 10 for normal use. (Default: ``6``)
+        rolloff (float, optional): The roll-off frequency of the filter, as a fraction of the Nyquist.
+            Lower values reduce anti-aliasing, but also reduce some of the highest frequencies. (Default: ``0.99``)
 
     Returns:
         Tensor: The waveform at the new frequency of dimension (..., time).
@@ -1386,7 +1394,7 @@ def resample(
     new_freq = new_freq // gcd
 
     kernel, width = _get_sinc_resample_kernel(orig_freq, new_freq, lowpass_filter_width,
-                                              waveform.device, waveform.dtype)
+                                              rolloff, waveform.device, waveform.dtype)
 
     num_wavs, length = waveform.shape
     waveform = torch.nn.functional.pad(waveform, (width, width + orig_freq))

torchaudio/transforms.py

@@ -640,16 +640,24 @@ class Resample(torch.nn.Module):
         orig_freq (float, optional): The original frequency of the signal. (Default: ``16000``)
         new_freq (float, optional): The desired frequency. (Default: ``16000``)
         resampling_method (str, optional): The resampling method. (Default: ``'sinc_interpolation'``)
+        lowpass_filter_width (int, optional): Controls the sharpness of the filter, more == sharper
+            but less efficient. We suggest around 4 to 10 for normal use. (Default: ``6``)
+        rolloff (float, optional): The roll-off frequency of the filter, as a fraction of the Nyquist.
+            Lower values reduce anti-aliasing, but also reduce some of the highest frequencies. (Default: ``0.99``)
     """
 
     def __init__(self,
                  orig_freq: int = 16000,
                  new_freq: int = 16000,
-                 resampling_method: str = 'sinc_interpolation') -> None:
+                 resampling_method: str = 'sinc_interpolation',
+                 lowpass_filter_width: int = 6,
+                 rolloff: float = 0.99) -> None:
         super(Resample, self).__init__()
         self.orig_freq = orig_freq
         self.new_freq = new_freq
         self.resampling_method = resampling_method
+        self.lowpass_filter_width = lowpass_filter_width
+        self.rolloff = rolloff
 
     def forward(self, waveform: Tensor) -> Tensor:
         r"""
@@ -660,7 +668,7 @@ class Resample(torch.nn.Module):
             Tensor: Output signal of dimension (..., time).
         """
         if self.resampling_method == 'sinc_interpolation':
-            return F.resample(waveform, self.orig_freq, self.new_freq)
+            return F.resample(waveform, self.orig_freq, self.new_freq, self.lowpass_filter_width, self.rolloff)
 
         raise ValueError('Invalid resampling method: {}'.format(self.resampling_method))