Jiang-Stan
Jiang-Stan
The .so file should be automatically generated by running setup.py file for installation. Please check if you have successfully installed Sparsebit.
> 请问你使用了多长时间的音频训练哈。能否上传一个演示音频呢？ 刚开始训，而且我这是有改动接了别的semantic token的。 按作者写的应该是在2k小时上训了100epoch吧，单机8卡 bs32
1. Yes 2. I successfully reproduce the result by uniformally generate 50 images from each class. Results is shown below(IS by torch-fidelity, others by [guided_diffusion evaluation code](https://github.com/openai/guided-diffusion/tree/main/evaluations). The paper claims...
> @pzelasko As for M subset, I am sure that I've called `.trim_to_supervisions` as I showed. I found the Supervisions available does not match with Feature available...  and it...
@SaltedSlark Hi, how long did you take preprocessing WenetSpeech M set? It takes me 50 minutes extracting features, but it has taken over 11 hours saving to `wenetspeech_cuts_M.jsonl.gz` and still...
I noticed that only one thread is set to save data from [here](https://github.com/lhotse-speech/lhotse/blob/master/lhotse/cut/set.py#L2295). I tried to use 32 threads but it still cannot finish saving. @pzelasko By separating recordings and...
> > I noticed that only one thread is set to save data from [here](https://github.com/lhotse-speech/lhotse/blob/master/lhotse/cut/set.py#L2295). I tried to use 32 threads but it still cannot finish saving. @pzelasko > >...
Can you share training detail of TTA autoencoderKL? I tried to reproduce the result with the origin repo with your preprocessed mel data and training configs, but I failed to...