Add model Wav2Letter (#462)

* add wav2letter model

* add unit_test to model

* add docstrings

* add documentation

* fix minor error, change logic on forward

* update padding same with ceil

* add inline typing and minor fixes to docstrings

* remove python2

* add formula do docstrings, change param name

* add test with mfcc, add pytest

* fix bug, update docstrings

* change parameter name

Author: tomassosorio

docs/source/models.rst

@@ -0,0 +1,17 @@
+.. role:: hidden
+    :class: hidden-section
+
+torchaudio.models
+======================
+
+.. currentmodule:: torchaudio.models
+
+The models subpackage contains definitions of models for addressing common audio tasks.
+
+
+:hidden:`Wav2Letter`
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: Wav2Letter
+
+  .. automethod:: forward

test/test_models.py

@@ -0,0 +1,30 @@
+import pytest
+
+import torch
+from torchaudio.models import Wav2Letter
+
+
+class TestWav2Letter:
+    @pytest.mark.parametrize('batch_size', [2])
+    @pytest.mark.parametrize('num_features', [1])
+    @pytest.mark.parametrize('num_classes', [40])
+    @pytest.mark.parametrize('input_length', [320])
+    def test_waveform(self, batch_size, num_features, num_classes, input_length):
+        model = Wav2Letter()
+
+        x = torch.rand(batch_size, num_features, input_length)
+        out = model(x)
+
+        assert out.size() == (batch_size, num_classes, 2)
+
+    @pytest.mark.parametrize('batch_size', [2])
+    @pytest.mark.parametrize('num_features', [13])
+    @pytest.mark.parametrize('num_classes', [40])
+    @pytest.mark.parametrize('input_length', [2])
+    def test_mfcc(self, batch_size, num_features, num_classes, input_length):
+        model = Wav2Letter(input_type="mfcc", num_features=13)
+
+        x = torch.rand(batch_size, num_features, input_length)
+        out = model(x)
+
+        assert out.size() == (batch_size, num_classes, 2)

torchaudio/models/__init__.py

@@ -0,0 +1 @@
+from .wav2letter import *

torchaudio/models/wav2letter.py

@@ -0,0 +1,74 @@
+from typing import Optional
+
+from torch import Tensor
+from torch import nn
+
+__all__ = ["Wav2Letter"]
+
+
+class Wav2Letter(nn.Module):
+    r"""Wav2Letter model architecture from the `"Wav2Letter: an End-to-End ConvNet-based Speech Recognition System"
+     <https://arxiv.org/abs/1609.03193>`_ paper.
+
+     :math:`\text{padding} = \frac{\text{ceil}(\text{kernel} - \text{stride})}{2}`
+
+    Args:
+        num_classes (int, optional): Number of classes to be classified. (Default: ``40``)
+        input_type (str, optional): Wav2Letter can use as input: ``waveform``, ``power_spectrum``
+         or ``mfcc`` (Default: ``waveform``).
+        num_features (int, optional): Number of input features that the network will receive (Default: ``1``).
+    """
+
+    def __init__(self, num_classes: int = 40,
+                 input_type: str = "waveform",
+                 num_features: int = 1) -> None:
+        super(Wav2Letter, self).__init__()
+
+        acoustic_num_features = 250 if input_type == "waveform" else num_features
+        acoustic_model = nn.Sequential(
+            nn.Conv1d(in_channels=acoustic_num_features, out_channels=250, kernel_size=48, stride=2, padding=23),
+            nn.ReLU(inplace=True),
+            nn.Conv1d(in_channels=250, out_channels=250, kernel_size=7, stride=1, padding=3),
+            nn.ReLU(inplace=True),
+            nn.Conv1d(in_channels=250, out_channels=250, kernel_size=7, stride=1, padding=3),
+            nn.ReLU(inplace=True),
+            nn.Conv1d(in_channels=250, out_channels=250, kernel_size=7, stride=1, padding=3),
+            nn.ReLU(inplace=True),
+            nn.Conv1d(in_channels=250, out_channels=250, kernel_size=7, stride=1, padding=3),
+            nn.ReLU(inplace=True),
+            nn.Conv1d(in_channels=250, out_channels=250, kernel_size=7, stride=1, padding=3),
+            nn.ReLU(inplace=True),
+            nn.Conv1d(in_channels=250, out_channels=250, kernel_size=7, stride=1, padding=3),
+            nn.ReLU(inplace=True),
+            nn.Conv1d(in_channels=250, out_channels=250, kernel_size=7, stride=1, padding=3),
+            nn.ReLU(inplace=True),
+            nn.Conv1d(in_channels=250, out_channels=2000, kernel_size=32, stride=1, padding=16),
+            nn.ReLU(inplace=True),
+            nn.Conv1d(in_channels=2000, out_channels=2000, kernel_size=1, stride=1, padding=0),
+            nn.ReLU(inplace=True),
+            nn.Conv1d(in_channels=2000, out_channels=num_classes, kernel_size=1, stride=1, padding=0),
+            nn.ReLU(inplace=True)
+        )
+
+        if input_type == "waveform":
+            waveform_model = nn.Sequential(
+                nn.Conv1d(in_channels=num_features, out_channels=250, kernel_size=250, stride=160, padding=45),
+                nn.ReLU(inplace=True)
+            )
+            self.acoustic_model = nn.Sequential(waveform_model, acoustic_model)
+
+        if input_type in ["power_spectrum", "mfcc"]:
+            self.acoustic_model = acoustic_model
+
+    def forward(self, x: Tensor) -> Tensor:
+        r"""
+        Args:
+            x (Tensor): Tensor of dimension (batch_size, num_features, input_length).
+
+        Returns:
+            Tensor: Predictor tensor of dimension (batch_size, number_of_classes, input_length).
+        """
+
+        x = self.acoustic_model(x)
+        x = nn.functional.log_softmax(x, dim=1)
+        return x