I tried to execute MozillaTTS directly by console in two different machines with Anaconda 3 and I got the same error message:
The first one is Centos 6.11 The second one is Virtual Machine Ubuntu 18.4 (host is Windows 10)
In both machines I have installed Python 3.8.10. When I execute a English sentence using the own tts model I got an very robotic voice. But when I execute the following sentence:
tts –text “Este es un ejemplo en castellano” --config_path config.json --model_path best_model.pth.tar --out_path OUTPUT/ejemplo.wav --vocoder_path vocoder_model.pth.tar --vocoder_config_path config_vocoder.json
I am using my own model (that I got by training) I have the following warning message:
[W NNPACK.cpp:79] Could not initialize NNPACK! Reason: Unsopported hardware.
and the following error message:
RuntimeError: Error(s) in loading state_dict for Tacotron2: size mismatch for embedding.weight: copying a param with shape torch.Size([129,512]) from checkpoint, the shape in current model is torch.Size([130,512]).
I don’t know what would be the problem because I installed TTS in some Windows 10 machines and I didn’t have problems.
Thanks a lot for your help