[css-speech-1] Useful values to drive TTS engines

As the goal of CSS Speech is to override default TTS engines behavior / values, one could ask why define normal, medium, and auto values? To define default/initial values and to block the inheritance of undesired custom values I suppose.

If yes, it could be interesting to mention it, when defining these three values, and to point, in each definition, to the concepts of inheritance and defaulting, as defined by CSS Cascading and Inheritance Level 4.

As matter of fact, all specialists of aural rendering are not CSS specialists. They may be puzzled by these values (as I was on first reading). Cascading is the most difficult part of CSS for people who have no experience in object programming.

normal is a value for:

• speak-as
• voice-stress
• voice-rate

medium is a value for:

• voice-pitch
• voice-range
• voice-rate
• voice-volume
• pause-before
• pause-after
• pause
• rest-before
• rest-after
• rest

auto is a value for:

• speak
• voice-duration

Start with the CSS Speech definitions for normal, medium, and auto values, then review the linked CSS Cascading and Inheritance Level 4 concepts of inheritance and defaulting. Done means the relevant value definitions explain their purpose and point readers to those concepts, covering the listed properties.