ym
> Hi, when I run > python synthesis.py --preset=presets/20180505_deepvoice3_ljspeech.json pretrained_models/20180505_deepvoice3_checkpoint_step000640000.pth sentences.txt pretrained_output > > error: > Command line args: > {'--checkpoint-postnet': None, > '--checkpoint-seq2seq': None, > '--file-name-suffix': '', > '--help':...
I make sure that in dataset part, I input all 10 classes name. Note that the two classes( nose, eyebrow) that not be detected are more smaller than other class,...
I see the english_characters , what about japanese? And too get the japanese_characters, token_type using is 'char' or 'bpe'? ENGLISH_CHARACTERS = [a-z],
The follow info may be helpful . ar = np.load(0_mel.npy) ar.shape ==> (80,48)
Any update? I also faiiled to convert BLIP model to TorchScript. > Hello, Is there any any to convert BLIP to TensorRT to have a lighter model on inference time...
> Hi @kaituoxu ! > I wanted to train tacotron2 on jsut. Can your implementation be trained on jsut? > I have the jsut preprocessed. I used deepvoice3's preprocess module!...
> I experimented with some changes to the base-template build pipeline before deciding it's too complicated and too far out of scope. The failed checks (the base image build and...
@lmxyy . "8-bit model support". Hope this can come soon and first. Because current INT4 flux combine with customed lora, can see a significant quality decline.
Agree! Training Refiner model is really important now if community want to get a good anime model like NAI. They must train a anime refiner, only base fine-tune is not...