AI voice cloning
AI Voice Cloning Online
Clone your voice from a clear 10–60 second recording, preview the result, and save it for text to speech. Membership required. Each preview uses 20 credits.
Hear the original and cloned voice
Listen to the original recording and the cloned voice. This sample is in Chinese.
Clone a voice in three steps
1
Upload a recording
Upload or record 10 to 60 seconds of clear speech. About 20 seconds of one speaker in a quiet room works best.
2
Generate a preview
Type a short line and hear the cloned voice. Each preview uses 20 credits, and failed previews are refunded.
3
Save and use it
Pick the saved voice in text to speech to generate voiceovers in your own voice. Saving a Fish Audio clone costs no credits.
Included with membership
- Clone models including Fish Audio and MiniMax
- Save multiple cloned voices to your account
- Monthly member credits for previews and text to speech
Only clone your own voice or a voice you have permission to use. Read the acceptable use policy first. Acceptable use policy
Frequently asked questions
Is voice cloning free?
Voice cloning is a member feature. A free account comes with signup credits to try text to speech and the voice library. Upgrade to clone your voice.
How many credits does a clone use?
Each preview uses 20 credits, and failed previews are refunded. Saving a Fish Audio clone costs no credits. Other models show the save cost before you confirm.
How long should the recording be?
Between 10 and 60 seconds. About 20 seconds of clear speech with little background noise gives the most stable clone.
Can I clone someone else's voice?
Only with their explicit permission. Before uploading, you confirm that you own the voice or are authorized to use it.
What can I do with a cloned voice?
Select it in text to speech to generate narration for videos, podcasts and audio content.