Why is it that even after training for over 70 epochs (766k steps), the generated audio in TensorBoard still isn't clear? #87
-
|
I trained for 766k steps using the same configuration but without SDP, yet the audio generated in TensorBoard is still unclear, with words being indistinct and muddled |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 2 replies
-
|
I cannot debug for free with so little information =) |
Beta Was this translation helpful? Give feedback.
-
|
anyway,I am still very grateful for the code you provided, which has been of great help to me! |
Beta Was this translation helpful? Give feedback.
-
|
Thanks, next time provide some more info so I or someone else can help you better. Good luck! =) |
Beta Was this translation helpful? Give feedback.
anyway,I am still very grateful for the code you provided, which has been of great help to me!