The audio-dataset-converter library allows the conversion between various dataset formats of audio datasets. Filters can be supplied as well, e.g., for cleaning up the data.
Dataset formats:
- classification: ADAMS (r/w), sub-dir (r/w), TXT (r/w)
- speech: ADAMS (r/w), CommonVoice (r/w), Festvox (r/w), Huggingface Audiofolder (r/w), TXT (r/w)