Document Model

This is the design of how a document is represented in memory and how the reader stream events are structured. This model is not an accurate representation of any specific document format, but is designed around the general structure and semantics of a document.

Fro example, an italic text span can be represented in many different ways in HTML, DocBook and other document formats. In the Cainteoir Text-to-Speech document model there is a single way of specifying an italic text span that is not tied to any specific document format.

The rationale behind this is that the consumers of the document model (text-to-speech engine, presentation layers, document converters, etc.) do not need to understand the different equivalent ways of specifying content. It also means the document model is a superset of all the different document formats.