Hello! Thank you for open-sourcing such a great project as RoboOmni. However, when I tried to reproduce the experiments on LIBERO-spatial, I found that the audio data was missing. I’m currently attempting to use Piper to convert the text prompts into audio, but it indicates that all test samples failed. Is there something wrong with the experimental approach I’m taking? I would really appreciate your guidance. Thank you!
Hello! Thank you for open-sourcing such a great project as RoboOmni. However, when I tried to reproduce the experiments on LIBERO-spatial, I found that the audio data was missing. I’m currently attempting to use Piper to convert the text prompts into audio, but it indicates that all test samples failed. Is there something wrong with the experimental approach I’m taking? I would really appreciate your guidance. Thank you!