w-okada / voice-changer

リアルタイムボイスチェンジャー Realtime Voice Changer
Other
16.65k stars 1.81k forks source link

VC PROCESSING EXCEPTION!!! index -1 is out of bounds for axis 0 with size 0 #595

Closed ajaxxxxxxx closed 1 year ago

ajaxxxxxxx commented 1 year ago

Issue Type

Bug Report

vc client version number

MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.10

OS

Windows 10

GPU

RTX 2070 Super

Clear setting

no

Sample model

yes

Input chunk num

no

Wait for a while

The GUI successfully launched.

read tutorial

yes

Extract files to a new folder.

yes

Voice Changer type

RVC

Model type

f0

Situation

Hello! Error I'm having is with a voice model I've been using for a few days now. For some reason, it works less often than not, and occasionally starts working again. When it doesn't work, it spews out this error in the console.

[Voice Changer] VC PROCESSING EXCEPTION!!! index -1 is out of bounds for axis 0 with size 0 Traceback (most recent call last): File "voice_changer\VoiceChanger.py", line 218, in on_request_sola File "voice_changer\RVC\RVC.py", line 176, in inference File "voice_changer\RVC\pipeline\Pipeline.py", line 169, in exec IndexError: index -1 is out of bounds for axis 0 with size 0

I'm not sure the cause, as other models work 100% fine all the time. This specific model only works after constantly restarting the program and readding the model to the program over and over, and eventually, works. I'd like to know if there's an actual solution.

w-okada commented 1 year ago

If other model work correctly, the model which can not be used might be an invalid model.

ajaxxxxxxx commented 1 year ago

If other model work correctly, the model which can not be used might be an invalid model.

As I'm typing this the model all of a sudden works. What would determine if the model is invalid?

ajaxxxxxxx commented 1 year ago

Update with more clarification: the model works fine when "index" is set to 0. Any value above 0 causes above error.

w-okada commented 1 year ago

So, the index is broken. Anyway, it sounds great that you can convert voice accurately. Please close this issue.