Implemented parameterized audio targets & hesitation.

[octavpo]

This implements issue #47.

[JackMostow]

Please explain what the hesitation part is. "Implemented hesitation" is far from self-explanatory.

[kevindeland]

None of the existing Bubble Pop activities work; they all show up as empty bubbles.

Also, I am guessing that this feature is supposed to be audio targets in BubblePop? Can you provide a sample data source so that the functionality can be tested?

[JackMostow]

@kevindeland - Yes, the audio targets are for BubblePop. I forget the syntax in the data source to specify whether bubbles are audio, labelled, or both. The audio and text labels typically match but needn't.
Audio-only bubbles are intentionally empty but colored.
Each audio target announces itself when it appears.

@octavpo, can you point to an example data source that uses audio targets?

@judithodili, now that we finally have this long-awaited feature, we can use it for its intended purposes, including phonological tasks with audio-only stimuli and targets, and text-to-speech mapping tasks in multiple choice form as an assessable alternative to oral reading. RoboTutor Development Plan lists several activities possible with audio targets:
1.1.2 see letter, pick spoken letter name
1.2.4 hear word, pick initial sound
1.2.5 see word, pick initial sound
1.3.2 see word, pick spoken word
See non-word, pick pronunciation (see Swahili pseudowords in list)
See numeral or quantity, pick spoken number (not listed; assessable alternative to speaking and reading numbers)
In fact audio target activities are multiple choice recognition tasks as assessable alternatives to almost any spoken-response tasks.

Question: Which such activities add most value, and who should create them? Note that many of them can use data sources that already exist, just by modifying the modality label.

[octavpo]

None of the existing Bubble Pop activities work; they all show up as empty bubbles.

Also, I am guessing that this feature is supposed to be audio targets in BubblePop? Can you provide a sample data source so that the functionality can be tested?

Sorry I should have put a reference to the issue where this feature is discussed, it #47. As it says there in order to use it you need to add properties "target_show" and "target_say" with values true or false (that's the default) to the data source files, similar to the current "question_show" and "question_say".

The reason you see empty bubbles is because I followed the way "question_show" and "question_say" are implemented, which both default to false. So since right now there are no "target_show" and/or "target_say" in the any of the data source files, you don't see the text and don't hear the targets either. I can change the default if we want to, but I think it would be better to keep consistency with "question_show" and "question_say".

I have rebased the branch on the current development branch and solved the conflicts, but this PR doesn't show it. Not sure how to make it see that. Maybe I need to close it and make a new one? It would be good to have these PRs merged sooner, so we can start from the latest code when we implement new features.