Add README to examples directory

examples/README.md

@@ -0,0 +1,12 @@
+The examples in this directory demonstrate applied use of the torchaudio library. They are included only for reference purposes.
+
+> [!NOTE]
+> **We have transitioned TorchAudio into a
+>  maintenance phase. Our main goals were to reduce redundancies with the rest of the
+>  PyTorch ecosystem, make it easier to maintain, and create a version of
+>  TorchAudio that is more tightly scoped to its strengths: processing audio
+>  data for ML. Because of this transition, the examples included here will no longer be maintained
+>  going forward. The scripts may not work properly as is. For hands on examples of features in the
+>  library that *are* still supported, see the "tutorials" subdirectory. Please see
+>  [our community message](https://github.com/pytorch/audio/issues/3902)
+>  for more details.**

examples/libtorchaudio/CMakeLists.txt

@@ -2,9 +2,6 @@ cmake_minimum_required(VERSION 3.5)
 
 project(libtorchaudio-cpp-example)
 
-SET(BUILD_SOX ON CACHE BOOL "Build libsox into libtorchaudio")
-
-SET(BUILD_KALDI OFF CACHE BOOL "Build Kaldi into libtorchaudio")
 SET(BUILD_RNNT ON CACHE BOOL "Build RNN transducer into libtorchaudio")
 SET(BUILD_TORCHAUDIO_PYTHON_EXTENSION OFF CACHE BOOL "Build Python binding")
 

examples/libtorchaudio/README.md

@@ -1,6 +1,5 @@
 # Libtorchaudio Examples
 
-* [Augmentation](./augmentation)
 * [Speech Recognition with wav2vec2.0](./speech_recognition)
 
 ## Build

examples/libtorchaudio/speech_recognition/README.md

@@ -1,10 +1,6 @@
 # Speech Recognition with wav2vec2.0
 
-This example demonstarates how you can use torchaudio's I/O features and models to run speech recognition in C++ application.
-
-**NOTE**
-This example uses `"sox_io"` backend for loading audio, which does not work on Windows. To make it work on
-Windows, you need to replace the part of loading audio and converting it to Tensor object.
+This example demonstarates how you can use torchaudio's models to run speech recognition in a C++ application.
 
 ## 1. Create a transcription pipeline TorchScript file
 

examples/self_supervised_learning/data_modules/__init__.py

@@ -1,4 +1,5 @@
 from ._hubert_datamodule import HuBERTDataModule
+from ._wav2vec2_datamodule import Wav2Vec2DataModule
 
 __all__ = [
     "HuBERTDataModule",