How do I clone a voice with AI?
On Kunya, go to Dashboard → Audio → Voice Cloning. Upload a clean audio sample (30 seconds to 3 minutes of the target voice), give it a name, and Kunya creates a custom voice model. You can then use that voice for text-to-speech anywhere in the platform.
Requirements for a good voice clone
- Clean audio — minimal background noise, music, or echo
- Duration — 30 seconds minimum; 1–3 minutes for best results
- Single speaker — no other voices in the sample
- Consistent delivery — varied sentence types (questions, statements, exclamations) improve range
Step by step
- Open Dashboard → Audio
- Click Voice Cloning or Custom Voices
- Upload your audio file (WAV or MP3, up to 20MB)
- Name your voice and click Create
- Use your voice in any TTS generation by selecting it from the voice picker
Consent and usage
Only clone voices you have the right to use. Kunya requires confirmation that you have consent from the voice owner before cloning.