dzr1026

Thank you very much for your work. Could you clarify whether the downstream task datasets are open-sourced? I would like to evaluate the capabilities of certain codecs. How can I...

Regarding the input for the downstream task

2

Thank you for your work. Could you please clarify what the input for the downstream task is? Is it the vector decoded by RVQ or the audio reconstructed by the...

About the input on ARCH benchmark

3

Thank you for your work, it's a very innovative piece of research. I have a question regarding the ARCH benchmark results (Table 5): What is the input for these results?...

The training data used for WavTokenizer-small

3

I would like to ask whether the dataset used to train WavTokenizer-small is the complete LibriTTS or the LibriTTS-580h mentioned in the paper. I am planning to reproduce this work...