Turn a manuscript into a casted audio production.

Import the manuscript

The author can begin by uploading an existing manuscript or entering text directly. Supported formats include PDF, DOCX, TXT, and HTML — a complete book can be uploaded at once. During project creation, the author provides the book title, author name, and default narrator voice.

The manuscript parser analyzes the document and separates it into chapters and passages. An optional setting allows quoted dialogue to be separated into individual passages so different characters can receive different voices. This creates an editable production structure without requiring the author to divide the manuscript manually.

Build the voice cast

Every project begins with a narrator. When additional speakers are detected during import, DreamCasters Studio creates cast positions for those characters so the author can review and assign their voices.

Detect and assign speakers

Inside each chapter, the author can select Auto Detect Speakers. The AI reviews the chapter passage by passage and attempts to determine which character is speaking. The system respects the cast the author has already configured — it uses existing cast names whenever the dialogue supports that assignment and introduces a new speaker only when a character does not already exist in the project cast.

Automatic detection accelerates the first casting pass, but the author remains in control. Any passage can still be reassigned manually.

Edit chapters and passages

Every chapter contains an ordered list of passages. For every passage, the author can change the assigned speaker, select a voice from the project cast or the complete voice catalog, edit the manuscript text, preview the assigned voice, narrate the individual passage, narrate it again after a change, and review the current audio status.

Add music and create the chapter mix

DreamCasters Studio includes a separate music library. Authors can upload MP3, WAV, and M4A files, organized into folders so projects with many songs, themes, ambient tracks, or chapter beds remain manageable. A music track can be assigned to a chapter, with control over music volume, fade in duration, fade out duration, track selection, and chapter assignment.

When the narration passages are ready, Mix With Music combines the rendered passage audio with the selected music bed into one continuous chapter WAV file. The completed chapter mixdown can be played inside the chapter workspace, saved to the project library, and downloaded as a WAV file — removing the need to export every passage and assemble the chapter manually in a separate editing application.

Render at any scale

Publish the audiobook

When the project is ready, the author selects Publish Book. The system creates a public listening page using the project identifier — no listener account or sign-in required. Listeners can view the book and chapter structure, play available chapters, move between chapters, and download completed WAV files. The author can remove public access at any time by selecting Unpublish, keeping control over when the audiobook is available and when it should be removed.

Deliver to the Listening Library

Published audio can also feed the DreamCasters Listening Library for registered members. Studio access requires a free DreamCasters Compass account. This separates the production environment from the listening experience: the author works inside DreamCasters Studio, public listeners use the published audiobook page, and registered members can access eligible work through the DreamCasters Listening Library.