Prepare a PDF or EPUB for AI narration
Prepare PDF and EPUB text for AI narration: check reading order, remove repeated headers, understand OCR and DRM limits, and test a short passage.
Review before you generate
Text extracted from a document may include repeated page headings, page numbers, broken words or unusual reading order. Review the extracted text before spending credits on narration.
PDFs need selectable text
ToAudio reads the text layer of a PDF. A scanned page is an image and needs OCR before this version can read it. Complex tables and multi-column layouts can lose their reading order, so copy and edit a short section first.
EPUB follows its reading order
A DRM-free EPUB has a spine describing chapter order. ToAudio follows that spine when extracting text. Encrypted books are not supported. Importing a book does not mean generating the entire book: select a passage of up to 4,000 characters for each narration request.
Listen to a short passage
Choose a voice and speaking pace, then generate a short passage containing names and difficult words. Correct your text before continuing with longer sections. The resulting voice is AI-generated.