[DOC] Standardization and minor fixes (#1892)

torchaudio/models/wavernn.py

@@ -283,7 +283,7 @@ class WaveRNN(nn.Module):
             specgram: the input spectrogram to the WaveRNN layer (n_batch, 1, n_freq, n_time)
 
         Return:
-            Tensor shape: (n_batch, 1, (n_time - kernel_size + 1) * hop_length, n_classes)
+            Tensor: shape (n_batch, 1, (n_time - kernel_size + 1) * hop_length, n_classes)
         """
 
         assert waveform.size(1) == 1, 'Require the input channel of waveform is 1'
@@ -343,7 +343,7 @@ class WaveRNN(nn.Module):
             lengths (Tensor or None, optional):
                 Indicates the valid length of each audio in the batch.
                 Shape: `(batch, )`.
-                When the ``specgram`` contains spectrograms with different duration,
+                When the ``specgram`` contains spectrograms with different durations,
                 by providing ``lengths`` argument, the model will compute
                 the corresponding valid output lengths.
                 If ``None``, it is assumed that all the audio in ``waveforms``
@@ -356,7 +356,7 @@ class WaveRNN(nn.Module):
                 1 stands for a single channel.
             Tensor or None
                 If ``lengths`` argument was provided, a Tensor of shape `(batch, )`
-                is retuned.
+                is returned.
                 It indicates the valid length in time axis of the output Tensor.
         """
 

torchaudio/pipelines/_tts/interface.py

@@ -25,7 +25,7 @@ class _TextProcessor(ABC):
             text (str or list of str): The input texts.
 
         Returns:
-            Tensor and Tensor:
+            (Tensor, Tensor):
             Tensor:
                 The encoded texts. Shape: `(batch, max length)`
             Tensor:
@@ -56,7 +56,7 @@ class _Vocoder(ABC):
                 The valid length of each sample in the batch. Shape: `(batch, )`.
 
         Returns:
-            Tensor and optional Tensor:
+            (Tensor, Optional[Tensor]):
             Tensor:
                 The generated waveform. Shape: `(batch, max length)`
             Tensor or None:

torchaudio/pipelines/_wav2vec2.py

@@ -134,7 +134,7 @@ class Wav2Vec2ASRBundle(Wav2Vec2Bundle):
             unk (str, optional): Token for unknown class. (default: ``'<unk>'``)
 
         Returns:
-            Tuple of strings:
+            Tuple[str]:
             For models fine-tuned on ASR, returns the tuple of strings representing
             the output class labels.
 

torchaudio/sox_effects/sox_effects.py

@@ -73,10 +73,10 @@ def apply_effects_tensor(
         sample_rate (int): Sample rate
         effects (List[List[str]]): List of effects.
         channels_first (bool, optional): Indicates if the input Tensor's dimension is
-            ``[channels, time]`` or ``[time, channels]``
+            `[channels, time]` or `[time, channels]`
 
     Returns:
-        Tuple[torch.Tensor, int]: Resulting Tensor and sample rate.
+        (Tensor, int): Resulting Tensor and sample rate.
         The resulting Tensor has the same ``dtype`` as the input Tensor, and
         the same channels order. The shape of the Tensor can be different based on the
         effects applied. Sample rate can also be different based on the effects applied.
@@ -191,20 +191,20 @@ def apply_effects_file(
             If input file is integer WAV, giving ``False`` will change the resulting Tensor type to
             integer type. This argument has no effect for formats other
             than integer WAV type.
-        channels_first (bool, optional): When True, the returned Tensor has dimension ``[channel, time]``.
-            Otherwise, the returned Tensor's dimension is ``[time, channel]``.
+        channels_first (bool, optional): When True, the returned Tensor has dimension `[channel, time]`.
+            Otherwise, the returned Tensor's dimension is `[time, channel]`.
         format (str or None, optional):
             Override the format detection with the given format.
             Providing the argument might help when libsox can not infer the format
             from header or extension,
 
     Returns:
-        Tuple[torch.Tensor, int]: Resulting Tensor and sample rate.
+        (Tensor, int): Resulting Tensor and sample rate.
         If ``normalize=True``, the resulting Tensor is always ``float32`` type.
         If ``normalize=False`` and the input audio file is of integer WAV file, then the
         resulting Tensor has corresponding integer type. (Note 24 bit integer type is not supported)
-        If ``channels_first=True``, the resulting Tensor has dimension ``[channel, time]``,
-        otherwise ``[time, channel]``.
+        If ``channels_first=True``, the resulting Tensor has dimension `[channel, time]`,
+        otherwise `[time, channel]`.
 
     Example - Basic usage
         >>>

torchaudio/transforms.py

@@ -787,7 +787,7 @@ class MuLawEncoding(torch.nn.Module):
             x (Tensor): A signal to be encoded.
 
         Returns:
-            x_mu (Tensor): An encoded signal.
+            Tensor: An encoded signal.
         """
         return F.mu_law_encoding(x, self.quantization_channels)
 
@@ -1629,7 +1629,7 @@ class PSD(torch.nn.Module):
                 of dimension `(..., channel, freq, time)` if multi_mask is ``True``
 
         Returns:
-            torch.Tensor: PSD matrix of the input STFT matrix.
+            Tensor: PSD matrix of the input STFT matrix.
                 Tensor of dimension `(..., freq, channel, channel)`
         """
         # outer product:
@@ -1773,7 +1773,7 @@ class MVDR(torch.nn.Module):
             eps (float, optional): a value added to the denominator in mask normalization. (Default: 1e-8)
 
         Returns:
-            torch.Tensor: the mvdr beamforming weight matrix
+            Tensor: the mvdr beamforming weight matrix
         """
         if self.multi_mask:
             # Averaging mask along channel dimension