Blakey Gavin

Single machine multi-gpus ddp training stuck

7

### Bug description At the link below I ran the code base and made almost no changes. [https://github.com/bytedance/music_source_separation](https://github.com/bytedance/music_source_separation) When I first ran the code for the above library, my NVIDIA...

make apex

2

When I use the "make apex" command to install apex, it cannot be installed, and then when I use "make apex.done" to install, the following error occurs:  How can...

class ISTFT(DFTBase): def __init__(self, n_fft=2048, hop_length=None, win_length=None, window='hann', center=True, pad_mode='reflect', freeze_parameters=True, onnx=False, frames_num=None, device=None): """PyTorch implementation of ISTFT with Conv1d. The function has the same output as librosa.istft. Args: n_fft:...

Hi，I would like to ask what good skills are there in training cyclegan？

替换 PESQ

3

您好，非常感谢您出色的工作。我想用你们的代码训练 48kHz 的语音模型，是否有可以代替 PESQ 的其它指标？ 我了解了一下 [visqol](https://github.com/google/visqol)，它两种模式（audio 和 speech），音频模式是 48kHz 的，但似乎不适用于语音。

Hello, may I ask if this model can control the duration of the generated music? Because I found that the duration of the generated music seems to be random. For...

ERROR: Exception in ASGI application

3

Hi, I encountered an error when running 'python app. py -- share' on Ubuntu 20.04. I hope to receive your help, thank you very much. The error is as follows:...

转换后音色跟着 source 而不是 target

10

您好，我将 ssl model 更换为中文版 wav2vec2 和 hubert，然后进行了 retrain 和 fine-tune，但不管哪种方式，转换出来的结果都是音色和 source 相似而不是 target。 请问可能的原因是什么，我应该怎么解决这个问题？

Hello, thank you very much for your outstanding work. Now, I would like to use it for my own dataset. If I were to use my own Chinese dataset (approximately...