Note If your dataset only consists of one column with audio, you can use a simple setup: store your audio files at the root.

Note or in a subdirectory

Note Multiple formats are supported at the same time, including AIFF, FLAC, MP3, OGG and WAV.

Note If you have several splits, you can put your audio files into directories named accordingly. This setup currently creates an extra unwanted label column (follow the bug resolution at https://github.com/huggingface/dataset-viewer/issues/3014).

Note If there is additional information you'd like to include about your dataset, like the transcription, add it as a `metadata.csv` file in your repository.

Note You can also use a JSONL file `metadata.jsonl`.

Note You can also locate the audio files in a different subdirectory of the split.

Note For audio classification datasets, you can also use a simple setup: use directories to name the audio classes.

Note You can also use provide multiple splits.

Note You can disable this automatic behavior in the YAML configuration. If your directory names have no special meaning, set `drop_labels: true` in the README header

Note The WebDataset format is well suited for large scale audio datasets. It consists of TAR archives containing audio files and their metadata and is optimized for streaming.