Create a project and add short script segments
Start a project and name the segments by their role: opening, first exchange, transition or closing. Each segment starts with free generation; optional templates provide suggested text for you to review and edit in the input box. This outline helps you see where a spoken explanation is doing too much work and where a short sound can carry the moment. You can refine the script before spending generation time. Read through the complete outline once to check that each segment earns its place in the episode. Each generation accepts up to 3,000 characters, including performance directions and selected character descriptions.
Use a consistent naming system for synthetic characters, then save and audition their voice samples. A segment can use up to three ready saved characters. Keep a note of each character’s timbre, language and performance direction, especially when the same person returns in a later scene. These descriptions are a useful creative reference throughout the project.
Choose a version by listening in context
Generation produces a clip for the selected segment. Listen to it on its own, then consider how it connects to the previous and next scenes. Compare available versions for clarity, tone and pacing rather than selecting the latest one automatically. Editing the script or cast makes existing audio stale, so generate a current version before including that changed segment in an export.
Arrange, review and take your audio with you
Place your chosen clips in numbered order and adjust volume, pauses and fades. Listen across joins, where timing can change the meaning of a line. Clips play in sequence; the studio has no separate background-music track or multitrack mixer. Export a WAV or MP3 episode, or a ZIP of processed individual WAV clips, the script and an arrangement manifest for another editor. Exports include pauses and can be up to 30 minutes. The model supports podcast content up to four minutes and other audio up to two minutes.