When installed onto a linux device, this program can be used as for home security analysing and classifiying live audio. This program listen to the live audio supplied by a microphone, splits the separate sound segments and classifies them against the testDogx classifier model. The results are sent using the PubNub api, which can be received by the supplied mobile app.
Installation commands
apt-get install python-opencv
apt-get install python-pygame
apt-get install lsof
apt-get install ipython python-opencv python-scipy python-numpy python-setuptools python-pip
apt-get install motion
apt-get install python-pyaudio python3-pyaudio
apt-get install python-dev
apt-get install python3-scipy
apt-get install libchromaprint-dev
apt-get install python-matplotlib
apt-get install libgsl0-dev
apt-get install python-sklearn
pip install pubnub
easy_install simplejson
wget http://sourceforge.net/projects/mlpy/files/mlpy%203.5.0/mlpy-3.5.0.tar.gz
tar xvf mlpy-3.5.0.tar.gz
cd mlpy-3.5.0
python setup.py install
apt-get install git
git clone https://github.com/tyiannak/pyAudioAnalysis.git
pip install scikits.talkbox
pip install eyeD3
apt-get install python-netifaces
- Install Dependencies.
- Download 'Project' Folder and place in /home.
- If placed elsewhere, change path in all scripts.
- Ensure input_device_index is correct in soundProcess.py, line 129.
- Ensure Motion is setup correctly and replace /etc/motion/motion.conf with supplied motion.conf file.
- Change the PubNub API keys to your own.
- Change IonicPush token keys inside PushPythonscript.py to your own.
- Or comment out the call in classifySounds.py, line 58.
- If using mobile app: set listenforapp.py to run on startup
Run soundProcess.py or set up mobile app to send the ON command.