Add transcript-reliant feature extractions to senselab

[900miles]

Description

Adds transcript-reliant feature extractions to senselab. Namely:

• average + variance length in seconds of given person's speech
• average + variance of words per minute of given person's speech
• mean length of utterance of given person's speech
• average + variance length in seconds of silence between turns / response latency
• mean length + variance of pauses mid-utterance

Unfortunately due to time constraints, this code hasn't been tested. There are some strong assumptions about the ScriptLine inputs that may not hold true in every case. There are also some feature extractions that might be better performed straight from audio / VAD output. This is a draft PR so there is a baseline of which to work from.

Related Issue(s)

#351

How Has This Been Tested?

This has not been tested.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Checklist:

• I have added tests to cover my changes.
• All new and existing tests passed.
• My code follows the code style of this project.