inconnu11
inconnu11
Did you normalize the Mel spectrogram? What's the range of the Mel spec?
> Hi @inconnu11 , > > In Keras `model.save` saves complete model and `model.save_weights` saves only weights of the model layers that includes frozen layers also.AFAIK, There is no direct...
> KerasNLP backbone Hi, my model is not KerasNLP backbone. I saved all the weights and re-assigned the lora parameters one by one.
Hi, I got it and thanks for the reply. But when I run the code with the default setting with LJSpeech corpus except toggle the type of prosody modelings to...
I'd like to do so. But it takes too long to train it. I have to train the model for 7days with one gpu T4. Are there any parts of...
Hi, I change the mdn loss calculation from fig1 to fig2 . But it doesn't seem to work. original MDN loss:  newer MDN loss: 
assistant_ids、user_ids、audio_ids还有两个说话人spk_0_ids、spk_1_ids的special token都是随机从文本token表里选择的吗
> just prepare some data with symbol 大概要多少小时的带标注数据啊