• I wrote this program for testing PocketSphinx and see the differencies with Apple Speech Recognition.

• An Artificial Intelligence Bot written in Python and Rivescript

• It has a command line mode and a speech mode

• For installation see 'requirements.txt' or read below

• Rivescript: pip install rivescript
• SpeechRecognition: Requirements: brew install portaudio , brew link portaudio , pip install pyaudio . then pip install SpeechRecognition
• Monotonic for python2: pip install monotonic
• Pocketsphinx: Requirements: brew install swig git . then pip install pocketsphinx
• (if you get the error 'OSError dlopen(libasound.so)' type the command 'csrutil disable' in recovery mode. this has worked for me.

• Type 'python main0.7.py' in Terminal
• The program asks for your Username (this name will be used to store your variables, ie. name, age, etc.).
• Then choose between command line mode or speech mode.
• Now the bot is ready to accept your input. Your chat history and your variables will be saved in a JSON file, so that once you log back (using the same Username) they will be restored and the bot will know it's you.

Inside the 'brain' folder you can customize the bot's knowledge and actions that correspond to your input.

• Most of the files in this folder are 'default', i just copied and pasted them, i only edited the 'begin' file so that it introduces to a new user asking for his name. I created the files 'system_functions.rive' and 'topics.rive'

• The basics system actions are controlling the input and output volume, empty trash, enable/disable wifi, list the wifi connection available.

• Please go to 'https://www.rivescript.com/docs/tutorial' for more info about Rivescript