Minsu Kang
Same error, too.. Did you solved the problem?
Try to run the example written in README.md. Is loss still 0?
> Hello, I tried to use a pretrained model with kss data, but a 1 second wav file was printed out. Can you only make a 1-second file with that...
Hi, @taewhankim . In this implementation, discriminator starts to train at 100,000 steps. Before 100,000 steps, "d" (discriminator loss) and "ad" (adversarial loss) are printed as zero. Also, I downsampled...
Hello, @chynphh 4 means the number of heads used to multihead-attention. If you edit the return value of multihead attention in pytorch, you can get the attention with (layer_num, head_num,...
Hi Hongpeng1992, You must change pytorch code as described in the README.md of this project. Please carefully see the section 2 of "FastSpeech" in README.md and change your pytorch code....
Hi, @ahmadsab95 . I tested on AMD Ryzen 3700x CPU and used 8 thread. It took within 8 minutes, but it may differs depending on what kind of computer components...
Dear @Eie1 , First of all, thank you for question. As you pointed out, there are some unnecessary steps in the tutorial code. If you already have grapheme-to-phoneme(g2p) mapping dictionary,...
Hi, @xueguoqing01 . I used a RTX-3090 GPU for all experiments.