The Audio creation page turns a written script into speech. Choose a supported speech model and voice, adjust the available delivery settings, then listen to the generated track before adding it to your project.
Write the words you want spoken
Open Audio and enter the exact script. Write spoken wording rather than a description of the video. For example, use “Meet your new everyday bottle” instead of “A narrator introduces the bottle.”
Check product names, numbers, punctuation and pronunciation-sensitive words. If you are translating a campaign, review the actual translated script instead of assuming a language setting will rewrite it for you.
Choose a model and voice
Open the model control beside the script, then choose a voice from that model's available list. Play a sample where one is provided. Some voices do not have a preview; you need to generate your script to hear that result.
Changing models changes the available voices and settings. Review the chosen voice again after a model switch, particularly when working in another language.
Shape the generated delivery
Open Settings to see the controls registered for the selected model. These may differ between models. Keep delivery instructions separate from words that should be spoken, and only use the controls the selected model supports.
The preview player's speed and volume change how you listen. They do not rewrite the saved generated track. Use an actual editing workflow if the delivered file needs a lasting timing or volume change.
Generate, listen and revise
- Review the script and the displayed generation estimate.
- Generate the track and wait for its result.
- Listen all the way through, especially names, pauses and the ending.
- Edit the script or supported voice settings and generate another take if needed.
Use the waveform and playback controls to inspect the result. Open your audio gallery to revisit a track, download it or reuse its script where available.
Use the track in a larger creation
Generated audio is available in the workspace library. You can attach an existing voice track to a saved actor, or bring audio into an editing workflow that accepts it. In Canvas, use the audio timeline controls to place and adjust a track alongside the design.
The current Audio page is a speech and voice-over tool. It does not provide a general music-generation studio, and a voice track alone does not create a lip-synced presenter video. Choose the appropriate connected Studio tools when your result requires those additional steps.
