hi @geneing ! I used the C++ code for infer! but the speed is so slowly
mels shape: (80, 500)
take times:53.759968996047974
Seven seconds of audio takes about 53 seconds!
my hparams
rnn_dims=256,
fc_dims=128,
sparsity_target=0.90,
How to optimize the params! thanks!
hi @geneing ! I used the C++ code for infer! but the speed is so slowly
mels shape: (80, 500)
take times:53.759968996047974
Seven seconds of audio takes about 53 seconds!
my hparams