This service is the main form of communication between the user and the application. It listens for a trigger phrase "Hey Chef" and then listen for an action to execute. This service also has an api to convert plain text into words to send to the speaker to then be transmited to the user
Compilation on Windows (and probably iOS): pip install SpeechRecognition pip install pyttsx pip install requests pip install Flask
Compilation and Setup on RPi: pip install SpeechRecognition
sudo apt-get install portaudio19-dev
sudo apt-get install libasound-dev
pip install pyaudio --user (needs to be 0.2.11)
pip install pyttsx --user
sudo apt-get install espeak
sudo apt-get install flac
pip install requests
pip install Flask
sudo apt-get install swig
#sudo apt-get install libpulse-dev
sudo apt-get install python-dev
sudo pip install pocketsphinx
sudo apt -y install python-pyaudio python3-pyaudio sox python3-pip python-pip libatlas-base-dev
SNOWBOY: Download tar file from PyPi, extract, and in extracted folder run "sudo python setup.py install"
FIX JACK ISSUE:
sudo apt-get install multimedia-jack
sudo dpkg-reconfigure -p high jackd2
#If "broken or not fully installed": first do "sudo apt-get install jackd2"
sudo adduser $(whoami) audio
After reboot: pulseaudio --kill
jack_control start
Audio Setup:
Go to: /usr/share/alsa/alsa.conf
- Change "0" to "1" in line: defaults.ctl.card 0
- Change "0" to "1" in line: defaults.pcm.card 0
Set audio output to USB device (Andrea Communications...)
Running: Testing: python commands_test.py
Import class: import NLP