jordand/

echo-embeddings-vctk-tar

Items: 109 This dataset ships as a single tar at the repo root. Members preserve paths like VCTK//audio.mp3 and VCTK//speaker_latent.safetensors. See loader.py for example loading. Attribution: Contains audio and embeddings derived from the CSTR VCTK Corpus. Distributed under CC BY 4.0; attribution required.