During training wavenet, predicted output audio is only of few sec (22kb) for every step.
sanbuddhacharyas opened this issue · comments
I have completed training tacotron model. Now, I am training wavenet there is no any problem in training but the evaluated output from wavenet is only of few sec (22kb) for each step. Why doesn't the model output's full audio?
you find any solution regarding this issue ? kindly let me know about your work and its progress .. leaving my email usmanbabar@gmail.com