search

xiph / LPCNet

Efficient neural speech synthesis
BSD 3-Clause "New" or "Revised" License
1.12k stars 295 forks source link

ValueError: all elements of `new_shape` must be non-negative #212

Open eklee0309 opened 2 months ago

eklee0309 commented 2 months ago

Hello.

I'm trying to test the trained model with training_tf2/test_lpcnet.py.

I got an error related to the numpy module,

"Traceback (most recent call last): File "test_lpcnet.py", line 72, in features = np.resize(features, (-1, nb_features)) ValueError: all elements of new_shape must be non-negative".

What do I have to do?

felixshing commented 1 month ago

same problem. Any solution?

Edit: okay, it turns out we should use reshape instead of resize...