haoheliu / AudioLDM-training-finetuning

AudioLDM training, finetuning, evaluation and inference.
https://audioldm.github.io/audioldm2/
MIT License
174 stars 34 forks source link

Inference is very slow #22

Open DucHuyAnalyst opened 7 months ago

DucHuyAnalyst commented 7 months ago

It takes about 2 minutes to create a 10-second audio

Screenshot 2023-12-04 160628