SpeechRecognizer#recognize_after_speech yields hypothesis instead of decoder about pocketsphinx-ruby

[watsonbox]

I've made a change so it should now be possible to do:

recognizer = Pocketsphinx::AudioFileSpeechRecognizer.new(configuration)
recognizer.recognize(file) do |hypothesis|
  puts hypothesis
  puts recognizer.decoder.words
end

You have to use the master branch to test this until I cut a new version.

[chinshr]

Yes, that works, thanks.