VeloTwin is your AI presenter: a voice clone and a face clone. In the app they are set up separately — Voice clone and Face clone in the sidebar — and combined when you add a VeloTwin to a video.
Accessing the Voice Library
Step 1: Select Voice clone from the left sidebar navigation menu. Step 2: Review the available options, organized into three sections:- Your Voices — custom voice clones you’ve created
- Workspace Wide Voices — a voice one of your teammates has promoted for the whole workspace to use (see below)
- Public Voices — a library of ready-to-use voices available to everyone
Sharing a voice with your workspace
Any custom voice you create can be promoted to Workspace Wide so every member of the workspace can use it — useful for a consistent narrator across a team’s videos. Once promoted, it appears under Workspace Wide Voices for all members.Setting Up a New Clone
Step 1: Click + Create new in the Your Voices section. Step 2: Enter a name for your voice clone in the text field (for example, “Andy”) and click Continue.
Recording the Calibration Script
Step 1: Choose your preferred microphone from the dropdown menu. Step 2: Click Start recording and read the provided script aloud. Step 3: Speak at a steady pace and maintain a relaxed, natural tone as you would in a normal conversation. This helps the system capture your actual sound so the clone feels natural. Step 4: Click Finish recording once you have completed the script.
Processing and Verification
After you finish recording, Velo will analyze the audio to clean up the sound and create your clone. If the system couldn’t capture your voice cleanly - often due to background noise or missed words - you will see a “Couldn’t capture your voice” notification. In this case, select Re-record to provide a clearer sample. Once the analysis is complete, your new voice clone will appear in your custom library and is ready for use in any future video projects.