Retrieval-based-Voice-Conversion-WebUI

Retrieval-based-Voice-Conversion-WebUI copied to clipboard

How to manually set validation samples?

1

I сouldn't find how an audio samples are divided into training and validation datasets. Any ideas?

kontomac

似乎是英文系统独有的BUG

4

运行go-realtime-gui.bat后会报错：OSError: [WinError 126] The specified module could not be found. Error loading "D:\RVC-beta-v2-0618\runtime\lib\site-packages\torch\lib\c10.dll" or one of its dependencies. 但是在中文系统上运行就没问题。 

yfw918

数据集质量检测

1

可以加入数据集质量检测功能，比如检测音域覆盖，元音辅音覆盖等（经常训练出来的模型某些发音发不出来或者是错的，比如yi发出来是e，in发出来是en），声音语言（可能需要区分？暂不清楚）

hpx502766238

如题，中英文数据集需要分开训练吗？为啥我训练出来的模型某些元音发音怪怪的，比如ou，o，有的元音直接发不出来，发o音出来yi音

hpx502766238

So i don't know anything about coding or ai things and it was so painful to install this on my computer. So i am assuming that i am missing some...

enescelik0

windows系统，用的是win一键包（5月份的版本）。使用v2模型进行推理时，可能会使输出音频时长被压缩，导致其音调偏高。更改了推理时的面板上的其他参数，这个问题依然存在。换成其他模型再换回来（或者说重新装载了这个模型），有可能使其恢复正常。我猜测这可能与输入音频的采样率有关，当时使用的输入音频采样率为44100Hz，出现这个问题时，被压缩后的音频时长与原音频时长之比接近400/441 。把输入音源重采样为40000Hz后，在用同一个模型进行推理，就能够正常使用，没有出现这个问题。

Wanlau

No module named 'faiss'

2

Hi, after installing the requirements I try to launch the web interface with python but I get the following error. `Traceback (most recent call last): File "infer-web.py", line 21, in...

Mshriver2

Start training can't know the progress %

1

Start training can't know the progress % Is there a way to know the current progress? The cmd show the 20% 50%... a lot of different progress, can not understand...

minimoonleaf

loading weights/TEST02.pth gin_channels: 256 self.spk_embed_dim: 109 Traceback (most recent call last): File "G:\RVC-beta-v2-0618\my_utils.py", line 14, in load_audio ffmpeg.input(file, threads=0) File "G:\RVC-beta-v2-0618\runtime\lib\site-packages\ffmpeg\_run.py", line 325, in run raise Error('ffmpeg', out, err) ffmpeg._run.Error:...

Wuxlol

Karmau-Git