This RVC model is trained over 200 epochs using a custom dataset consisting of around 40 minutes of audio clips extracted from Sezen Aksu's well-known songs. It is intended for generating realistic voice imitations of Sezen Aksu for the purpose of creating music covers. The model's training was conducted using Colab with the RVC Disconnected V2 framework, and Tensorboard was utilized for monitoring the training process.
If you have any questions, feel free to contact me on Discord: "kr4ken."