TPdfSpokenText - PDFiumPas

Съхранява низа, подаден на speech engine, плюс span-ове, които го съпоставят обратно с текста на PDF страницата

Syntax

type TPdfSpokenText = record ... end;

Field	Type	Purpose
`Text`	`WString`	String passed to the speech engine
`Spans`	`TPdfSpokenSpans`	Mapped runs that translate spoken offsets back to page text

A TPdfSpokenText chunk can represent a page, one reading unit, or any caller-defined read-aloud segment

API names, type names, file names, and PDF terms keep their original spelling so code search remains reliable
Build chunks with SpokenTextAppend or BuildSpokenTextFromWords