AI voiceovers and voice cloning

The challenge
Users want a faster and more flexible way to add narrated audio tracks to their Mozaik video projects without needing to record voiceovers manually or use external text-to-speech tools.

While a separate feedback item already covers adding general voiceover tracks to video projects — including recording a voiceover directly in Mozaik or uploading an existing audio file — this request focuses specifically on AI-generated voiceovers and voice cloning.

Related feedback item: https://feedback.mozaik-app.com/p/add-voiceover-tracks-to-video-projects

Our approach
Enable users to generate AI voiceovers directly inside the Mozaik video editor using modern text-to-speech technology. Users should be able to enter a script, generate a voiceover audio track, and add that generated audio directly to their video project.

This should support two main AI voice workflows:

• 1. Select from existing AI voices: Users should be able to choose from a library of AI voices and select a matching voice based on the use case, tone, language, target audience, or type of video they are creating.

• 2. Clone their own voice: Users should be able to clone their own voice and add it to the voice library of their Mozaik account, so they can generate future voiceovers using their own reusable AI voice clone.

The generated voiceover should behave like a normal audio track inside the editor, so users can place it in the video timeline and use it as narration for tutorials, explainer videos, social media clips, and other narrated formats.

Expected impact
• Faster creation of narrated videos without external recording or TTS tools
• Easier production of tutorials, explainers, and social media clips
• More flexibility through AI voice selection by use case, tone, language, or video type
• Stronger personalization through account-specific voice clones
• A more complete voiceover workflow when combined with the existing general voiceover-track request