Search is not available for this dataset
audio audioduration (s) 1.33 5 |
|---|
End of preview. Expand in Data Studio
This dataset is sampled from LibriVox audio (135+ hours). The purpose of this dataset is for training neural audio codecs or autoencoder to reconstruct high-fidelity audio data.
- Downloads last month
- 62