Huelloween processes an audio stream (microphone, wav recording) and adjusts Phillips Hue lights for that scary Halloween effect! It analyzes the amplitude of each frequency using the [Fast Fourier Transform] (https://en.wikipedia.org/wiki/Fast_Fourier_transform) (or FFT) algorithm.
Edit the config.yaml file, setting your Phillips Hue bridge IP address.
Run Huelloween accepting the mic as the audio input.
Test using the microphone for audio input; first get the audio devices and set in config file
python huelloween.py --list-audio-devices
python huelloween.py --test-micTest using a sample file as input
python huelloween.py --test-soundRun normally
python huelloween.pyThe first time you run - you may get prompted to "pair" your program or machine with the Phillips Hue Bridge. For example:
Press button on Bridge then hit Enter to try again
Attempting to set up your Phillips Hue Bridge
INFO:phue:Error opening config file, will attempt bridge registration
INFO:phue:Writing configuration file to /Users/mark/.python_hue
INFO:phue:Reconnecting to the bridge
INFO:phue:Attempting to connect to the bridge...
INFO:phue:Using ip: 10.0.1.2
In order to get PyAudio working I had to first install pyaudio C lib. I used Macports:
brew install portaudio
sudo port install portaudio
-
Phue Philips Hue Python Library providing a simple, effective and up to date library.
-
Juliana Pena for your helpful article showing just how to process audio in Python and applying the FFT properly!
-
Sounds taken from SoundBible
The purpose of Huelloween started just before Halloween 2014. I wanted to mimic the audio processing of scary recordings to adjust my Phillips Hue lightbulbs. Each window on the front side of the house would have a light bulb. It ended up being five windows in all with a speaker sitting just outside in a dark spot. Some children were scared, parents loved it, while I am sure my neighbors were happy to see it up for only one night.