aac-datasets’s documentation

Audio Captioning unofficial datasets source code for AudioCaps, Clotho, and MACS, designed for PyTorch.

• Installation
  • Python requirements
  • External requirements (AudioCaps only)
• Usage
  • Download a dataset
  • Load data
  • Build PyTorch DataLoader
  • About AudioCaps download
• About datasets subsets
  • AudioCaps
  • Clotho
  • MACS
  • WavCaps
  • Datasets overlaps
• API
  • aac_datasets package