Release 1.1.1

Updates from the previous release:

• Listening back to an annotation (label) now clears the overlay
• Performance optimisations within the detection pipeline
• "Completed?" command now also recognises "complete?"
• Answers are now sanitised to remove quiz activation keywords and quiz response words so a feedback loop can't begin
• Activation phrases and answers are more leniently detected now (more likely to "hear" you correctly)
• iPads now lock their orientation and camera orientation to portrait
• Quiz questions can now be repeated with another asking of "quiz me"
• Updated commands and added technologies to README

Known issues:

• If no audio activity (commands provided) occurs over a consecutive minute, Apple's speech-to-text technology idles out and stops using the microphone. The app then needs to be restarted. To remedy this in a future release, you would record the Unix time (or another datetime format) of the last command to be heard. Then you would run a timer that makes a callback every x seconds (say, 5 seconds) to see if the time since the last command is near a minute. If so, you would restart the SpeechRecogniser (call self.recognizer.resetTranscript() in the view controller).