Add audio mixing into data model and TTML representation

[nigelmegitt]

We say surprisingly little at the moment about how to define audio mixing and link in audio resources. It's technically there via the profile definition, but it needs to be added to the data model and the TTML representation so it's clear how to use it.

[nigelmegitt]

Planning for this:

• Add Audio to data model
• Add Mixing Instruction to data model
• Define Audio as either an <audio> with a src attribute, as a child of a Text (<p>) or a TTS directive on a Text, being tta:speak or tta:pitch
• clipBegin and clipEnd to be supported
• xml:lang to be supported? If so, must match xml:lang of parent
• Define Mixing Instruction as a static value of tta:pan or tta:gain or a list of <animate> elements that vary those attributes (to support fill="freeze") - can be added to Script Events, Texts and Audios.
• Add examples
• Add informative explanation of audio mixing semantics - maybe use something like https://w3c.github.io/adpt/#webaudio-section ?
• Make sure that profile feature requirements are complete and adequate for audio

[nigelmegitt]

While working on the specification for adding audio recordings I reminded myself of the various ways in which an audio recording can be embedded and referenced, of which there are at least 5 in total. Requirement R15 of DAPT is clear that both referenced and embedded options need to be available, but should we be syntactically restricting the options for each? Will raise as separate issues.