Huge delay and lags on good GPU

[pasta-pasta]

Issue Type

Bug Report

vc client version number

MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.11

OS

Windows 10

GPU

RTX 3060 12gb

Clear setting

no

Sample model

no

Input chunk num

yes

Wait for a while

The GUI successfully launched.

read tutorial

yes

Extract files to a new folder.

yes

Voice Changer type

RVC

Model type

pyTorch

Situation

Im trying to use voicechanger on my pc. The configuration is a bit lame, i have i5-4460 with 32gb of ddr3 RAM and RTX 3060 with 12 gb of VRAM.

After launching the voicechanger, the delay between my words and converted ones are 10-20 seconds, and they are extremely laggy. On any chunk size.

I'm guessing that there is a problem in my internet connection, bc my router died and i'm using my phone usb modem to share internet, but im not sure.

application window capture

logs on terminal

Preparing your application. Please wait... (node:14896) electron: Failed to load URL: http://localhost:18888/ with error: ERR_CONNECTION_REFUSED Booting PHASE :main Booting PHASE :MMVCServerSIO [Voice Changer] model slot is changed -1 -> 3 ................RVC -102 ERR_CONNECTION_REFUSED http://localhost:18888/

[w-okada]

Is there anything running in the background?

Please check the Windows Performance Monitor to see how the load is being managed.

[pasta-pasta]

nothing serious. And on the first tests there were only two browsers present without all the other programs.

[w-okada]

try

https://github.com/w-okada/voice-changer/blob/master/tutorials/trouble_shoot_communication_ja.md

with audio file

[pasta-pasta]

somehow using models that are not preset fixes the problem. i will check default models later when i'll upgrade to new CPU. thx for help!