Haohe Liu / 刘濠赫
Haohe Liu / 刘濠赫
@thorwhalen Looks like the checkpoint file is corrupted. May be that's because you have interrupted once when AudioLDM tried to download the checkpoint. Please try to delete the corrupt checkpoint...
Sorry, we current do not have plans to put the fine-tuned model on to hugging face because we found it performs not so well. You are welcome to contribute though.
This indicates the download of the model checkpoint has once been interrupted by either you or the system. Please delete all the files in the ~/.cache/audioldm path and try again.
Hi please refer to this issue https://github.com/haoheliu/AudioLDM/issues/9
Another similar issue here https://github.com/haoheliu/AudioLDM/issues/4
Retraining with 44.1 is totally feasible. A typical GPU can do this job. Our next stage of research will address the higher sampling rate.
@Jiraxys Hi thanks for reporting this. I'll look into it now.
@Jiraxys Hi now the issue should be fixed. Please re-clone this project and try again. Thanks!
Thanks for your advice. OK I will work on it soon. When I have done it i will reach u out.
> Colab notebook implementation please! I implement a demo code for subband analysis and synthesis on colab: [code on colab](https://colab.research.google.com/drive/1n2ed4ucPEqTTV4MVa-FXyLzeDweBIoex?usp=sharing)