Get All Available Datasets in Torchaudio

Torchaudio provides a collection of datasets, which can be incredibly useful for several reasons. It can be used for training and evaluation of audio models. The datasets serve as a valuable resource for research and development, enabling the exploration of new ideas, algorithms, and advancements in audio processing. This tutorial explains how to get all available datasets in Torchaudio.

In the following code, we retrieve a list of all available datasets in Torchaudio. We iterate over a list and print the name of each dataset to the console.

import torchaudio

datasets = torchaudio.datasets.__all__

for name in datasets:
    print(name)

Here's an example of what the output might look like:

COMMONVOICE
LIBRISPEECH
LibriLightLimited
SPEECHCOMMANDS
VCTK_092
...