Yumevox

Voice Synthesis

Clone and synthesize voices with high fidelity.

.pth Upload

Upload Model

.index Upload

Upload Index

Reference Audio

Upload Audio

Pitch Shift

Index Rate

0.3

Volume Envelope

1.0

Protect

0.33

F0 Method

Prepare your audio data for training

Training Audio Files

Upload Audio Files

📝 Tip: Upload multiple audio files (10+ minutes total) of the same speaker for best results. Quality matters more than quantity!

Model Name

Sample Rate

CPU Cores

F0 Method

Enable Noise Reduction

Apply Audio Effects

Train your custom voice model

Select Preprocessed Model

Total Epochs

Save Every Epoch

Batch Size

GPU ID

Enable Overtraining Detection

Save Only Latest