Please help me in preparing data step

[cod3r0k]

Hi, I have an Arabic dataset structured as follows:

mydata/wavs/*.wav       # Audio files (22,050 Hz)
mydata/metadata.csv     # Metadata file in the format: id|transcription

I'm unsure about the next steps.

Previously, I used Coqui and eSpeak, which handled everything automatically. But now I'm working with the vits_pytorch code, and I'm stuck at Step 2 — generating mel spectrograms. I'm not sure what I need to do at this stage.

Could you guide me on how to proceed?