Voice Studio
Give every character their own unique voice — designed by you!
How it works
Step 1 — Design the voice
Describe what the character sounds like and click "Hear the Voice" to try it out. Keep adjusting until it sounds just right!
Step 2 — Generate training audio
The computer reads a special script in the character's voice for about 10 minutes. This becomes the "lesson" for the next step.
Step 3 — Train the voice model
The computer studies that audio and learns the character's voice. This takes about 2 hours — great to run overnight!
Step 4 — Perform & convert
Record yourself reading the character's lines in your own voice, then the computer transforms it into the character's voice!
Setup Required
Voicebox (voice design + TTS)
git clone https://github.com/jamiepine/voicebox
cd voicebox/backend
pip install -r requirements.txt
python ../server.py --host 127.0.0.1 --port 8000
Then open Voicebox in your browser to design character voices before coming back here.
RVC (voice conversion + training)
git clone https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI
cd Retrieval-based-Voice-Conversion-WebUI
pip install -r requirements.txt
python infer-web.py --port 7865
