RVC v2 Voice-to-Voice Conversion
RVC v2 converts your spoken voice into any trained target voice while preserving tone, emotion and delivery, with a dataset builder to create new voices.
- Rvc Project
- 2 variants
- From $0.0002475 per processing second
- Price
- See price
- Speed
- ~15s
- Best for
- Voice to Text
- Price
- See price
- Speed
- See speed
- Best for
- Voice · Voice to Voice
Working with RVC.
RVC v2 is voice conversion rather than text-to-speech: you speak, and the model re-renders your performance in a trained target voice, keeping your tone, emotion and timing. The dataset variant automates building an RVC v2 training set from a URL.
When to choose RVC
Choose RVC when the performance matters and you want to keep your own delivery in someone else's voice. XTTS is the pick for cloning a voice from a short sample; Mureka Vocal Clone is the singing-focused option.
Choosing between the variants
RVC v2 is the conversion model and needs a trained target voice. RVC Dataset is the preparation step that builds a training set automatically from a URL.
Questions about RVC
The short answers. For anything specific to your setup, talk to our team.
Talk to an engineerPlan for your expected volume.You supply a spoken performance rather than text, so timing, emotion and emphasis come from your recording instead of being invented.
It automatically assembles an RVC v2 voice cloning dataset from a URL, which is the input a voice model needs before training.
Yes — tone, emotion and natural delivery carry over from your input voice into the target voice.