CorentinJ / Real-Time-Voice-Cloning

Clone a voice in 5 seconds to generate arbitrary speech in real-time
Other
52.05k stars 8.71k forks source link

Missing synthesizer pretrained.pt #948

Closed AsterTheWanderer closed 2 years ago

AsterTheWanderer commented 2 years ago

I already tried these files https://drive.google.com/drive/folders/1aPYBbabQGNFHp6DegcKhfwNyLTXcKg5f?usp=sharing from Francisco https://drive.google.com/drive/folders/1lb-LlS8Sx9RqcGzuV6GxvKHk-PC9TqQx?usp=sharing from Alex https://drive.google.com/file/d/1n1sPXvT34yXFLT47QZA6FIRGrwMeSsZc/view from RobbeW

I still get the FileNotFoundError: [Errno 2] No such file or directory: 'synthesizer\\saved_models\\pretrained\\pretrained.pt' any help?

AsterTheWanderer commented 2 years ago

it's already been fixed

AsterTheWanderer commented 2 years ago

Installation overly simplified by Aster

Step 1

all you need to do is download the whole folder in the google drive it's the whole repo and the models are already included so you don't need to download it separately.

Link: https://drive.google.com/drive/folders/1couiJINgIAaT3R73_i1rkIFnSxl7reOh?usp=sharing

Step 2

You MUST run it using python 3.6 or 3.7 run cd command to the folder

Example: cd C:\Users\Asus\Downloads\Real-Time-Voice-Cloning-master

Step 3

check your numpy ver. You MUST have numpy ver below 1.20

if you're using conda and doesn't have pip yet run conda install pip

if you have numpy ver above 1.20 run command pip uninstall numpy then run the command pip install numpy==1.20 after uninstalling the numpy

after that install pytorch (https://pytorch.org/get-started/locally/)

pick in these options in the pytorch website PyTorch Build = Stable (1.10) Your OS = [pick the OS that you're using] Package = [pick the package you're using ex. Conda, pip, LibTorch, or Source] Language = Python Compute Platform = CUDA 10.2 Run this Command = [a command should appear here. The command that appears here depends on your choices]

copy the command that appeared in the 'Run this command' then run it

Example: conda install pytorch torchvision torchaudio cudatoolkit=10.2 -c pytorch

Step 4

install ffmpeg run pip install ffmpeg

Step 5

run pip install -r requirements.txt to install the remaining necessary packages.

Step 6 (completely optional)

run python demo_cli.py to check if there's any error. If there's none you're good to go.

Step 7

run the program run the command python demo_toolbox.py

Then you're done it should've run by now.