Create a Cloned Voice from an Audio Sample

Use voice cloning to create a reusable voice based on an audio recording. This is useful for retakes, voice continuity, or creating variations of the same voice with different tone or emotion.

Overview

A cloned voice is created from a clean audio sample of a speaker.
Once created, the voice can be reused across tasks and projects, depending on access settings.

Step 1. Open the Voices page

Go to Voices: https://studio.dubformer.ai/voices

Step 2. Start creating a new voice

Click Add new voice clone.

Step 3. Configure the voice clone

Fill in the required fields:

  • Folder — select where the cloned voice will be stored.

  • Voice name — enter a clear and descriptive name.

  • Voice style — choose the appropriate style for the voice.

  • Audio sample — upload an audio recording.

Audio requirements:

  • Minimum length: 60 seconds

  • Maximum length: 2 minutes

  • Clean speech only

  • Minimal background noise, echo, or distortion

Step 4. Create the clone

Click Add clone.
The system will generate a cloned voice based on the provided sample.

Step 5. Choose the cloning mode Expressive or Standard

Step 6. Use the voice

The cloned voice becomes available in the Voice selection, Prooflistening, and Lip-sync tasks under the Speakers tab.