As mentioned in the Expert connect with Roberto. I'd like to know in advance during content development what words or utterances are (a) not yet unsupported (b) are banned or "illegal" e.g. curses or any unacceptable words (child related laws?)
One idea is to provide a tool to enable devs to stage and validate content that is spoken or expected to be recognized by Jibo. A secondary and important feature is that devs have a staging area for their interactive content for the entire skill. It helps to document what's going on, supports TEAM collaboration, and also supports future translations. It should also ignore markup tags so that we stage all aspects of the animations, voice, etc. That way it is easier to see the whole "play". Please include a special comment tag also - director's notes, etc;) From the staging area, one can copy past to mims or build content rules. Devs might use comments to show where the content used, etc.
I like the idea of having a content building tool where one can write out a detailed "script" for the whole script. I do this anyways but with a staging area we can then validate it and it documents the skill centrally. From this speech rules and Mims can be built or updated.
This is a bit more detailed than the kscript mentioned by Andrew and presented in the docs. My background includes theatre, so I started with my desktop app (Srcivener) for writing scripts/plays to get a feel for big picture. The trouble is we will not really be able to test our words and uterrances without a live robot.... and of course, with live testing, it will not be effecient and I think we can improve the "design phase" by Jibo speaking the staged content... so for my last feature idea:
--> perhaps via the jibo SDK, Jibo can speak the whole text from the staging area text tool. Here devs might then better sense the dialog flow, tailor their voice timings, modifications,etc, etc. As suggested with kscripts, devs could first play the whole script through the robot as part of the design phase, get the content in good shape and then provide the underlying skill implementation.
future: the staging area might evolve to provide more debugging support (on markup tags,etc) or even referenceable content blocks for the Mims... ah, stop me typing...ok.