ericwudayi

No, but maybe we can try to solve it. Does it make sense that we feed the melspectrogram output generated by pretrained autoencoder as training data of MelGAN?

Hi @dy2009 , This is a very cool and useful feature! Hoping can open source it also! I found the input feature of hifidecoder is gpt latent space, but not...

Actually, I run all the workflow testing file by replacing the execution function with newer one, `execution_with_graph`. I pass almost all the tests except for following nine. I take a...

Sure, in flyteidl the state is type of ExecutionState. ``` enum ExecutionState { // By default, all executions are considered active. EXECUTION_ACTIVE = 0; // Archived executions are no longer...

Use enum class make the usage not so straight-forward. User need to declare ExecutionState when update. But it is very clear. ``` from ... import ExecutionState remote.update_execution( ..., state=ExecutionState.EXECUTION_ARCHIVED, tags=["..."])...

我是使用vits-fast-finetuning 的chinese-to-ipa来实作的. 比较繁复的点是如何让ipa跟char对齐. 但这个问题我后来发现跟ipa没有关系. 问题是我在data_util.py 没有加上bert.zero_...